(19)
(11)EP 3 265 823 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
06.05.2020 Bulletin 2020/19

(21)Application number: 16710793.7

(22)Date of filing:  07.03.2016
(51)Int. Cl.: 
G01N 33/68  (2006.01)
(86)International application number:
PCT/GB2016/050626
(87)International publication number:
WO 2016/142696 (15.09.2016 Gazette  2016/37)

(54)

AMBIENT IONIZATION MASS SPECTROMETRY IMAGING PLATFORM FOR DIRECT MAPPING FROM BULK TISSUE

BILDGEBUNGSPLATTFORM FÜR UMGEBUNGSIONISIERUNGSMASSENSPEKTROMETRIE FÜR DIREKTE ABBILDUNG AUS GESAMTGEWEBE

PLATE-FORME D'IMAGERIE PAR SPECTROMÉTRIE DE MASSE À IONISATION AMBIANTE POUR LE MAPPAGE DIRECT DE TISSU VOLUMINEUX


(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(30)Priority: 06.03.2015 GB 201503876
06.03.2015 GB 201503879
06.03.2015 GB 201503864
06.03.2015 GB 201503877
06.03.2015 GB 201503867
06.03.2015 GB 201503863
06.03.2015 GB 201503878
09.09.2015 GB 201516003
16.10.2015 GB 201518369

(43)Date of publication of application:
10.01.2018 Bulletin 2018/02

(73)Proprietor: Micromass UK Limited
Wilmslow SK9 4AX (GB)

(72)Inventors:
  • JONES, Emrys
    Manchester M20 6AW (GB)
  • PRINGLE, Steven Derek
    Hoddlesden Darwen BB3 3PS (GB)
  • TAKÁTS, Zoltán
    Haslingfield Cambridge CB23 1JE (GB)

(74)Representative: Dehns 
St. Bride's House 10 Salisbury Square
London EC4Y 8JD
London EC4Y 8JD (GB)


(56)References cited: : 
  
  • J. BALOG ET AL: "Intraoperative Tissue Identification Using Rapid Evaporative Ionization Mass Spectrometry", SCIENCE TRANSLATIONAL MEDICINE, vol. 5, no. 194, 17 July 2013 (2013-07-17) , pages 194ra93-194ra93, XP055271114, Washington, DC ISSN: 1946-6234, DOI: 10.1126/scitranslmed.3005623
  • OTTMAR GOLF ET AL: "Rapid Evaporative Ionization Mass Spectrometry Imaging Platform for Direct Mapping from Bulk Tissue and Bacterial Growth Media", ANALYTICAL CHEMISTRY, vol. 87, no. 5, 3 March 2015 (2015-03-03), pages 2527-2534, XP055272752, ISSN: 0003-2700, DOI: 10.1021/ac5046752
  • ROSSITZA LAZOVA ET AL: "Imaging Mass Spectrometry-A New and Promising Method to Differentiate Spitz Nevi From Spitzoid Malignant Melanomas", AMERICAN JOURNAL OF DERMATOPATHOLOGY, vol. 34, no. 1, 1 February 2012 (2012-02-01), pages 82-90, XP055273544, US ISSN: 0193-1091, DOI: 10.1097/DAD.0b013e31823df1e2
  • Karl-Christian Schäfer ET AL: "In Situ, Real-Time Identification of Biological Tissues by Ultraviolet and Infrared Laser Desorption Ionization Mass Spectrometry", Analytical Chemistry, vol. 83, no. 5, 1 March 2011 (2011-03-01), pages 1632-1640, XP55047007, ISSN: 0003-2700, DOI: 10.1021/ac102613m
  
Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


Description

FIELD OF THE INVENTION



[0001] The present invention generally relates to mass spectrometry, and in particular to methods of ion imaging, methods of electrosurgery, ion imagers, mass spectrometers and electrosurgical devices. Various embodiments are contemplated wherein analyte ions generated by an ambient ionisation ion source are then subjected either to: (i) mass analysis by a mass analyser such as a quadrupole mass analyser or a Time of Flight mass analyser; (ii) ion mobility analysis (IMS) and/or differential ion mobility analysis (DMA) and/or Field Asymmetric Ion Mobility Spectrometry (FAIMS) analysis; and/or (iii) a combination of firstly ion mobility analysis (IMS) and/or differential ion mobility analysis (DMA) and/or Field Asymmetric Ion Mobility Spectrometry (FAIMS) analysis followed by secondly mass analysis by a mass analyser such as a quadrupole mass analyser or a Time of Flight mass analyser (or vice versa). Various embodiments also relate to an ion mobility spectrometer and/or mass analyser and a method of ion mobility spectrometry and/or method of mass analysis.

BACKGROUND



[0002] Mass spectrometry imaging ("MSI") analysis of biological samples is known and allows simultaneous and spatially resolved detection of metabolites, proteins and lipids directly from biological tissue sections.

[0003] The technique has gained significant momentum during the course of the last two decades with the introduction of new techniques such as matrix assisted laser desorption/ionization ("MALDI"), secondary ion mass spectrometry ("SIMS") and desorption electrospray ionization ("DESI").

[0004] The spatially resolved nature of the resulting data allows its use as a supplemental layer of information for histopathological characterization and classification of tissues including the possibility of cancer biomarker discovery.

[0005] Rapid evaporative ionization mass spectrometry ("REIMS") may be used for the real time identification of tissues e.g., during surgical interventions. Coupling of mass spectrometry with a surgical diathermy device has resulted in a technology known as intelligent knife ("iKnife") technology which has an intra-operative tissue identification accuracy of 92-100 %.

[0006] iKnife technology allows surgeons to more efficiently resect tumours intraoperatively through minimizing the amount of healthy tissue removed whilst ensuring that all the cancerous tissue is removed.

[0007] Rapid evaporative ionization mass spectrometry analysis of biological tissue has been shown to yield phospholipid profiles showing high histological and histopathological specificity - similar to Matrix Assisted Laser Desorption Ionisation ("MALDI"), Secondary Ion Mass Spectrometry ("SIMS") and Desorption Electrospray lonisation ("DESI") imaging. A mass spectrometric signal is obtained by subjecting the cellular biomass to alternating electric current at radiofrequency which causes localized Joule-heating and the disruption of cells along with desorption of charged and neutral particles. The resulting aerosol or surgical smoke is then transported to a mass spectrometer and/or ion mobility spectrometer for on-line mass spectrometric and/or ion mobility analysis.

[0008] Conventional rapid evaporative ionization mass spectrometry profiling applications require a spectral library of reference mass spectra in order to build multivariate classification models which are necessary for pattern-based identification.

[0009] Current iKnife technology reference mass spectra are obtained by manual electrosurgical sampling of ex vivo tissue specimens followed by the histopathological examination of the remaining material. Although the conventional workflow provides satisfactory data, there is a degree of uncertainty involved at the validation step since the tissue part producing the spectral data cannot be investigated since it is evaporated during the course of the analysis. Hence, conventionally all identifications are based on interpolation of the histological environment of the evaporated tissue.

[0010] J. Balog et al., Sci. Transl. Med. 5(194), 93 (2013) is concerned with intraoperative tissue identification using rapid evaporative ionization mass spectrometry.

[0011] O. Golf et al., Anal. Chem. 87(5), 2527 (2015) is concerned with a rapid evaporative ionization mass spectrometry imaging platform for direct mapping from bulk tissue and bacterial growth media.

[0012] R. Lazova et al., Am. J. Dermatopathol. 34(1), 82 (2012) is concerned with imaging mass spectrometry to differentiate Spitz nevi from Spitzoid malignant melanomas.

[0013] K. Shafer et al., Anal. Chem. 83(5), 1632 (2011) is concerned with in situ, real-time identification of biological tissues by ultraviolet and infrared laser desorption ionization mass spectrometry.

[0014] It is desired to provide an improved method of ion imaging.

SUMMARY



[0015] According to an aspect there is provided a method of ion imaging as claimed in claim 1.

[0016] In contrast to the known manual data collection approach, exemplary embodiments relate to an automated computer-controlled method of ambient ionization mass spectrometry sampling of tissue specimens wherein the 3D tissue environment may be used for histological validation.

[0017] In some embodiments, an ambient ionization mass spectrometry imaging device may be used in a minimally invasive fashion for the analysis of macroscopic tissue slices (not histological sections) and both the adjacent slice and the remaining tissue material may be fixed, embedded, sectioned, stained and histologically examined.

[0018] Although the very cells giving the spectral data may still be evaporated, the complete 3-dimensional adjacent environment gives sufficient information about their histological classification. Exemplary embodiments provide an imaging platform for systematic ambient ionization mass spectrometry data and/or ion mobility data collection which can serve as a basis for iKnife technology applications.

[0019] Further embodiments provide a mass spectrometric imaging platform for sample preparation-free ambient imaging MS analysis of biological samples.

[0020] Rapid evaporation ionization mass spectrometry ("REIMS") technology allows real time intra-operative tissue classification. In order to create spectral libraries for training the classification models, reference data needs to be acquired in large quantities as classification accuracy generally improves as a function of number of training samples.

[0021] Various aspects provide automated high-throughput methods for collecting ambient ionization mass spectrometry data and/or ion mobility data from heterogeneous organic tissue.

[0022] In exemplary embodiments, the instrumentation includes a 2D stage with an additional high-precision z-axis actuator which may be equipped with an electrosurgical diathermy-based sampling probe.

[0023] The sample may include a biological sample, a biological tissue, human tissue, animal tissue biological matter, a bacterial colony, a fungal colony or one or more bacterial strains. In general, the method may comprise a non-surgical or non-therapeutic method of ion imaging.

[0024] The sample can comprise native or unmodified sample material.

[0025] The native or unmodified sample material may be unmodified by the addition of a matrix or reagent.

[0026] The biological tissue may comprise in vivo biological tissue, ex vivo biological tissue or in vitro biological tissue.

[0027] The biological tissue comprises either: (i) adrenal gland tissue, appendix tissue, bladder tissue, bone, bowel tissue, brain tissue, breast tissue, bronchi, coronal tissue, ear tissue, esophagus tissue, eye tissue, gall bladder tissue, genital tissue, heart tissue, hypothalamus tissue, kidney tissue, large intestine tissue, intestinal tissue, larynx tissue, liver tissue, lung tissue, lymph nodes, mouth tissue, nose tissue, pancreatic tissue, parathyroid gland tissue, pituitary gland tissue, prostate tissue, rectal tissue, salivary gland tissue, skeletal muscle tissue, skin tissue, small intestine tissue, spinal cord, spleen tissue, stomach tissue, thymus gland tissue, trachea tissue, thyroid tissue, ureter tissue, urethra tissue, soft and connective tissue, peritoneal tissue, blood vessel tissue and/or fat tissue; (ii) grade I, grade II, grade III or grade IV cancerous tissue; (iii) metastatic cancerous tissue; (iv) mixed grade cancerous tissue; (v) a sub-grade cancerous tissue; (vi) healthy or normal tissue; or (vii) cancerous or abnormal tissue.

[0028] The sample classification model may include a biological sample classification model, a biological tissue classification model, a human tissue classification model, an animal tissue classification model, a biological matter classification model, a bacterial colony classification model, a fungal colony classification model or a bacterial strain classification model.

[0029] Constructing, training or improving the sample classification model may be in order either: (i) to distinguish between healthy and diseased tissue; (ii) to distinguish between potentially cancerous and non-cancerous tissue; (iii) to distinguish between different types or grades of cancerous tissue; (iv) to distinguish between different types or classes of sample material; (v) to determine whether or not one or more desired or undesired substances are present in the sample; (vi) to confirm the identity or authenticity of the sample; (vii) to determine whether or not one or more impurities, illegal substances or undesired substances are present in the sample; (viii) to determine whether a human or animal patient is at an increased risk of suffering an adverse outcome; (ix) to make or assist in the making a diagnosis or prognosis; and (x) to inform a surgeon, nurse, medic or robot of a medical, surgical or diagnostic outcome.

[0030] Using the obtained mass spectral data and/or ion mobility data to construct, train or improve the sample classification model may comprise performing a supervised or unsupervised multivariate statistical analysis of the mass spectral data and/or ion mobility data.

[0031] The multivariate statistical analysis may be selected from the group consisting of: (i) principal component analysis ("PCA"); and (ii) linear discriminant analysis ("LDA").

[0032] The method may further comprise analysing a profile of the aerosol, smoke or vapour or a profile of ions derived from the aerosol, smoke or vapour.

[0033] The profile may be selected from the group consisting of: (i) a lipidomic profile; (ii) a fatty acid profile; (iii) a phospholipid profile; (iv) a phosphatidic acid (PA) profile; (v) a phosphatidylethanolamine (PE) profile; (vi) a phosphatidylglycerol (PG) profile; (vii) a phosphatidylserines (PS) profile; (viii) a phosphatidylinositol (PI) profile; or (ix) a triglyceride (TG) profile.

[0034] The first device may comprise or form part of an ambient ion or ionization source or the first device may generate the aerosol, smoke or vapour for subsequent ionization by an ambient ion or ionization source or other ionization source.

[0035] The first device may be arranged and adapted to generate aerosol, smoke or vapour from the sample without the sample requiring prior preparation.

[0036] The first device may comprise an ion source selected from the group consisting of: (i) a laser desorption ionization ("LDI") ion source; (ii) a laser diode thermal desorption ("LDTD") ion source; and (iii) a laser ablation electrospray ("LAESI") ion source.

[0037] The aerosol, smoke or vapour may comprise uncharged aqueous droplets optionally comprising cellular material.

[0038] At least 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90% or 95% of the mass or matter generated by the first device and which forms the aerosol may be in the form of droplets.

[0039] The first device may be arranged and adapted to generate aerosol wherein the Sauter mean diameter ("SMD", d32) of the aerosol is in a range: (i) < 5 µm; (ii) 5-10 µm; (iii) 10-15 µm; (iv) 15-20 µm; (v) 20-25 µm; or (vi) > 25 µm.

[0040] The aerosol may traverse a flow region with a Reynolds number (Re) in the range: (i) < 2000; (ii) 2000-2500; (iii) 2500-3000; (iv) 3000-3500; (v) 3500-4000; or (vi) > 4000.

[0041] Substantially at the point of generating the aerosol, the aerosol may comprise droplets having a Weber number (We) selected from the group consisting of: (i) < 50; (ii) 50-100; (iii) 100-150; (iv) 150-200; (v) 200-250;(vi) 250-300; (vii) 300-350; (viii) 350-400; (ix) 400-450; (x) 450-500; (xi) 500-550; (xii) 550-600; (xiii) 600-650; (xiv) 650-700; (xv) 700-750; (xvi) 750-800; (xvii) 800-850; (xviii) 850-900; (xix) 900-950; (xx) 950-1000; and (xxi) > 1000.

[0042] Substantially at the point of generating the aerosol, the aerosol may comprise droplets having a Stokes number (Sk) in the range: (i) 1-5; (ii) 5-10; (iii) 10-15; (iv) 15-20; (v) 20-25; (vi) 25-30; (vii) 30-35; (viii) 35-40; (ix) 40-45; (x) 45-50; and (xi) > 50.

[0043] Substantially at the point of generating the aerosol, the aerosol may comprise droplets having a mean axial velocity selected from the group consisting of: (i) < 20 m/s; (ii) 20-30 m/s; (iii) 30-40 m/s; (iv) 40-50 m/s; (v) 50-60 m/s; (vi) 60-70 m/s; (vii) 70-80 m/s; (viii) 80-90 m/s; (ix) 90-100 m/s; (x) 100-110 m/s; (xi) 110-120 m/s; (xii) 120-130 m/s; (xiii) 130-140 m/s; (xiv) 140-150 m/s; and (xv) > 150 m/s.

[0044] The first device may comprise a point of care ("POC"), diagnostic or surgical device.

[0045] Exemplary methods include aspirating the aerosol, smoke or vapour produced from the sample. In some embodiments, the method may further include aspirating the aerosol, smoke or vapour in a substantially pulsed, discontinuous or irregular manner. In some embodiments, the method may further include varying an aspiration duty cycle during the course of a surgical, non-surgical or other procedure.

[0046] In some embodiments, the method may further include heating the collision surface.

[0047] The step of heating the collision surface may include heating the collision surface to a temperature selected from the group consisting of: (i) < about 100 °C; (ii) about 100-200 °C; (iii) about 200-300 °C; (iv) about 300-400 °C; (v) about 400-500 °C; (vi) about 500-600 °C; (vii) about 600-700 °C; (viii) about 700-800 °C; (ix) about 800-900 °C; (x) about 900-1000 °C; (xi) about 1000-1100 °C; and (xii) > about 1100 °C.

[0048] In exemplary embodiments, the method can also include mass analysing and/or ion mobility analysing the analyte ions in order to obtain the mass spectral data and/or ion mobility data corresponding to each location.

[0049] The method may further comprise mass analysing and/or ion mobility analysing the aerosol, smoke or vapour or ions derived from the aerosol, smoke or vapour in order to obtain the mass spectral data and/or ion mobility data corresponding to each location.

[0050] Various embodiments are contemplated wherein analyte ions generated by an ambient ionisation ion source are then subjected either to: (i) mass analysis by a mass analyser such as a quadrupole mass analyser or a Time of Flight mass analyser; (ii) ion mobility analysis (IMS) and/or differential ion mobility analysis (DMA) and/or Field Asymmetric Ion Mobility Spectrometry (FAIMS) analysis; and/or (iii) a combination of firstly ion mobility analysis (IMS) and/or differential ion mobility analysis (DMA) and/or Field Asymmetric Ion Mobility Spectrometry (FAIMS) analysis followed by secondly mass analysis by a mass analyser such as a quadrupole mass analyser or a Time of Flight mass analyser (or vice versa). Various embodiments also relate to an ion mobility spectrometer and/or mass analyser and a method of ion mobility spectrometry and/or method of mass analysis.

[0051] In exemplary embodiments, the method may further include operating the first device in a cutting mode of operation. In such embodiments, the first device may form one or more substantially continuous cuts in the sample. In some embodiments, the method may further include maintaining the first device at substantially the same height over the sample whilst performing the one or more substantially continuous cuts in the sample. In some embodiments, the method may further include operating the first device in a pointing mode of operation.

[0052] In exemplary embodiments, the method may further include obtaining an optical image of the sample. In some embodiments, the method may further include substantially co-registering the optical image and an ion image. In some embodiments, the method may further include defining one or more regions of interest in the optical image and/or the ion image. The method can include, in some embodiments, determining a class or classification of one or more regions of interest. For example, the class or classification may include a healthy status, a pre-cancerous status, a cancerous status or a bacterial strain.

[0053] An embodiment provides a method that includes sampling a plurality of different locations of a sample using a first device arranged and adapted to generate aerosol, smoke or vapour from the sample to obtain mass spectral data and/or ion mobility data at each location, and using a sample classification model which was previously constructed, trained or improved according to a method of ion imaging as described above in order to classify the sample at each location.

[0054] Using the obtained mass spectral data and/or ion mobility data to construct, train or improve a sample classification model may comprise analysing one or more sample spectra of said mass spectral data and/or ion mobility data so as to classify an aerosol, smoke or vapour sample.

[0055] Analysing the one or more sample spectra so as to classify the aerosol, smoke or vapour sample may comprise supervised analysis of the one or more sample spectra and/or unsupervised analysis of the one or more sample spectra.

[0056] Analysing the one or more sample spectra so as to classify the aerosol, smoke or vapour sample may comprise using one or more of: univariate analysis; multivariate analysis; principal component analysis (PCA); linear discriminant analysis (LDA); maximum margin criteria (MMC); library-based analysis; soft independent modelling of class analogy (SIMCA); factor analysis (FA); recursive partitioning (decision trees); random forests; independent component analysis (ICA); partial least squares discriminant analysis (PLS-DA); orthogonal (partial least squares) projections to latent structures (OPLS); OPLS discriminant analysis (OPLS-DA); support vector machines (SVM); (artificial) neural networks; multilayer perceptron; radial basis function (RBF) networks; Bayesian analysis; cluster analysis; a kernelized method; and subspace discriminant analysis.

[0057] Analysing the one or more sample spectra so as to classify the aerosol, smoke or vapour sample may comprise developing a classification model or library using one or more reference sample spectra.

[0058] Analysing the one or more sample spectra so as to classify the aerosol, smoke or vapour sample may comprise performing linear discriminant analysis (LDA) after performing principal component analysis (PCA).

[0059] Analysing the one or more sample spectra so as to classify the aerosol, smoke or vapour sample may comprise performing a maximum margin criteria (MMC) process after performing principal component analysis (PCA).

[0060] Analysing the one or more sample spectra so as to classify the aerosol, smoke or vapour sample may comprise defining one or more classes within a classification model or library.

[0061] Analysing the one or more sample spectra so as to classify the aerosol, smoke or vapour sample may comprise defining one or more classes within a classification model or library manually or automatically according to one or more class or cluster criteria.

[0062] The one or more class or cluster criteria for each class may be based on one or more of: a distance between one or more pairs of reference points for reference sample spectra within a model space; a variance value between groups of reference points for reference sample spectra within a model space; and a variance value within a group of reference points for reference sample spectra within a model space.

[0063] The one or more classes may each be defined by one or more class definitions.

[0064] The one or more class definitions may comprise one or more of: a set of one or more reference points for reference sample spectra, values, boundaries, lines, planes, hyperplanes, variances, volumes, Voronoi cells, and/or positions, within a model space; and one or more positions within a class hierarchy.

[0065] Analysing the one or more sample spectra so as to classify the aerosol, smoke or vapour sample may comprise using a classification model or library to classify one or more unknown sample spectra.

[0066] Analysing the one or more sample spectra so as to classify the aerosol, smoke or vapour sample may comprise classifying one or more sample spectra manually or automatically according to one or more classification criteria.

[0067] The one or more classification criteria may comprise one or more of:

a distance between one or more projected sample points for one or more sample spectra within a model space and a set of one or more reference points for one or more reference sample spectra, values, boundaries, lines, planes, hyperplanes, volumes, Voronoi cells, or positions, within the model space being below a distance threshold or being the lowest such distance;

a position for one or more projected sample points for one or more sample spectra within a model space being one side or other of one or more reference points for one or more reference sample spectra, values, boundaries, lines, planes, hyperplanes, or positions, within the model space;

a position for one or more projected sample points for one or more sample spectra within a model space being within one or more volumes or Voronoi cells within the model space; and

a probability or classification score being above a probability or classification score threshold or being the highest such probability or classification score.



[0068] Various embodiments are contemplated which relate to generating smoke, aerosol or vapour from a sample or a target (details of which are provided elsewhere herein) using an ambient ionisation ion source. The aerosol, smoke or vapour is then mixed with a matrix and aspirated into a vacuum chamber of a mass spectrometer and/or ion mobility spectrometer. The mixture is caused to impact upon a collision surface causing the aerosol, smoke or vapour to be ionised by impact ionization which results in the generation of analyte ions. The resulting analyte ions (or fragment or product ions derived from the analyte ions) may then be mass analysed and/or ion mobility analysed and the resulting mass spectrometric data and/or ion mobility spectrometric data may be subjected to multivariate analysis or other mathematical treatment in order to determine one or more properties of the sample or the target in real time.

[0069] The laser device may be a non-contact surgical device.

[0070] A non-contact surgical device may be defined as a surgical device arranged and adapted to dissect, fragment, liquefy, aspirate, fulgurate or otherwise disrupt biologic tissue without physically contacting the tissue.

[0071] As the non-contact device may not make physical contact with the tissue, the procedure may be seen as relatively safe and can be used to treat delicate tissue having low intracellular bonds, such as skin or fat.

[0072] According to various embodiments the mass spectrometer and/or ion mobility spectrometer may obtain data in negative ion mode only, positive ion mode only, or in both positive and negative ion modes. Positive ion mode spectrometric data may be combined or concatanated with negative ion mode spectrometric data. Negative ion mode can provide particularly useful spectra for classifying aerosol, smoke or vapour samples, such as aerosol, smoke or vapour samples from samples or targets comprising lipids.

[0073] Ion mobility spectrometric data may be obtained using different ion mobility drift gases, or dopants may be added to the drift gas to induce a change in drift time of one or more species. This data may then be combined or concatenated.

[0074] It will be apparent that the requirement to add a matrix or a reagent directly to a sample may prevent the ability to perform in vivo analysis of tissue and also, more generally, prevents the ability to provide a rapid simple analysis of sample or target material.

BRIEF DESCRIPTION OF THE DRAWINGS



[0075] Various embodiments will now be described, by way of example only, and with reference to the accompanying drawings in which:

Fig. 1 illustrates a method of rapid evaporative ionization mass spectrometry ("REIMS") wherein an RF voltage is applied to bipolar forceps resulting in the generation of an aerosol or surgical plume which is captured through an irrigation port of the bipolar forceps and is then transferred to a mass spectrometer and/or ion mobility spectrometer for ionization and mass analysis and/or ion mobility analysis;

Fig. 2 shows a rapid evaporative ionization mass spectrometry imaging platform located above a tissue sample to be imaged;

Fig. 3 shows a workflow of a combined Desorption Electrospray lonisation ("DESI") and rapid evaporative ionization mass spectrometry imaging platform analysis for co-registration of histological features between an optical image and Desorption Electrospray lonisation ("DESI") and rapid evaporative ionization mass spectrometry data;

Fig. 4 shows a heated coil interface used on a Waters Xevo G2-S (RTM) instrument for improved sensitivity and robustness towards contamination;

Fig. 5 shows a method of analysis that comprises building a classification model according to various embodiments;

Fig. 6 shows a set of reference sample spectra obtained from two classes of known reference samples;

Fig. 7 shows a multivariate space having three dimensions defined by intensity axes, wherein the multivariate space comprises plural reference points, each reference point corresponding to a set of three peak intensity values derived from a reference sample spectrum;

Fig. 8 shows a general relationship between cumulative variance and number of components of a PCA model;

Fig. 9 shows a PCA space having two dimensions defined by principal component axes, wherein the PCA space comprises plural transformed reference points or scores, each transformed reference point or score corresponding to a reference point of Fig. 7;

Fig. 10 shows a PCA-LDA space having a single dimension or axis, wherein the LDA is performed based on the PCA space of Fig. 9, the PCA-LDA space comprising plural further transformed reference points or class scores, each further transformed reference point or class score corresponding to a transformed reference point or score of Fig. 9;

Fig. 11 shows a method of analysis that comprises using a classification model according to various embodiments;

Fig. 12 shows a sample spectrum obtained from an unknown sample;

Fig. 13 shows the PCA-LDA space of Fig. 10, wherein the PCA-LDA space further comprises a PCA-LDA projected sample point derived from the peak intensity values of the sample spectrum of Fig. 12

Fig. 14 shows a method of analysis that comprises building a classification library according to various embodiments;

Fig. 15 shows a method of analysis that comprises using a classification library according to various embodiments;

Fig. 16 schematically illustrates a setup of rapid evaporative ionization mass spectrometry imaging instrumentation;

Fig. 17A shows a rapid evaporative ionization mass spectrometry imaging sampling probe, Fig. 17B shows a variety of possible alternatively shaped sampling probes and Fig. 17C shows a setup of a xyz-stage wherein the sampling probe is mounted onto a z-actuator and is connected to a high voltage power supply and wherein evaporated aerosol is captured by suction tubing and is transported to a mass spectrometer;

Fig. 18A shows a cutting sampling mode of the rapid evaporative ionization mass spectrometry imaging platform and Fig. 18B shows a pointing sampling mode of the rapid evaporative ionization mass spectrometry imaging platform;

Fig. 19A shows the impact on carbonization, burning-valley and crater size for various cutting speeds in a cutting mode of operation, Fig. 19B shows the impact on carbonization, burning-valley and crater size for the time the electrosurgical tip remained inside the sample for a pointing mode of operation and Fig. 19C shows how rapid evaporative ionization mass spectrometry imaging in cutting mode at low spatial resolution evaporates the top surface layer of the sample;

Fig. 20A shows concordance correlation coefficients (CCC) between rapid evaporative ionization mass spectrometry imaging in a cutting mode of operation and iKnife technology mass spectra in dependency on varying frequency at 2 kV for porcine liver and Fig. 20B shows concordance correlation coefficients (CCC) between rapid evaporative ionization mass spectrometry imaging in a cutting mode of operation and iKnife technology mass spectra in dependency on varying voltage at 40 kHz for porcine liver;

Fig. 21A shows total ion counts (TIC) at different frequencies in a cutting mode of operation and Fig. 21B shows total ion counts (TIC) at different voltages in a cutting mode of operation;

Fig. 22A shows concordance correlation coefficients between rapid evaporative ionization mass spectrometry imaging in a pointing mode of operation and iKnife technology mass spectra in dependency on frequency, Fig. 22B shows concordance correlation coefficients (CCC) between rapid evaporative ionization mass spectrometry imaging in a pointing mode of operation and iKnife technology mass spectra in dependency on voltage, Fig. 22C shows total ion counts (TIC) at different frequencies in a pointing mode of operation and Fig. 22D shows total ion counts (TIC) at different voltages in a cutting mode of operation;

Fig. 23A shows a mass spectral pattern of porcine liver obtained in a cutting mode of operation for high voltages, Fig. 23B shows a mass spectral pattern of porcine liver obtained in a cutting mode of operation for low voltages and Fig. 23C shows an iKnife technology reference spectrum;

Fig. 24 shows a principal component analysis plot of various kinds of tissue types analysed with the same experimental rapid evaporative ionization mass spectrometry imaging parameters for cutting and pointing modes respectively;

Fig. 25 shows a sample, H&E and mass spectrometric multivariate images of liver samples with metastatic tumour analysed by rapid evaporative ionization mass spectrometry and Desorption Electrospray lonisation ("DESI") wherein it is apparent that both techniques clearly differentiate the tissue types;

Fig. 26 shows principal component analysis plots of healthy and cancerous liver tissues for rapid evaporative ionization mass spectrometry imaging cutting and pointing modes as well as for Desorption Electrospray lonisation ("DESI") data wherein PC is the principal component and percentage values are explained variance;

Fig. 27 shows an univariate intensity comparison of single phospholipid ion species wherein the depicted images of samples are ion-images of the respective ions and Desorption Electrospray lonisation ("DESI") and rapid evaporative ionization mass spectrometry show similar relative intensity values for the same ions wherein PE is phosphatidyl-ethanolamine;

Fig. 28 shows an example profile for S. pneumoniae acquired using a modified Tecan EVO (RTM) without the introduction of isopropanol (IPA) matrix;

Fig. 29A shows spectral profiles obtained for Fusobacterium nucleatum using the Tecan platform without IPA, Fig. 29B shows spectral profiles obtained for Fusobacterium nucleatum using forceps without IPA and Fig. 29C shows spectral profiles obtained for Fusobacterium nucleatum using forceps with IPA;

Fig. 30A shows spectral profiles obtained for Staphylococcus hominis using automated Tecan based rapid evaporative ionization mass spectrometry without IPA, Fig. 30B shows spectral profiles obtained for Staphylococcus hominis using forceps based rapid evaporative ionization mass spectrometry without IPA, Fig. 30C shows spectral profiles obtained for Staphylococcus hominis using automated Tecan based rapid evaporative ionization mass spectrometry with IPA and Fig. 30D shows spectral profiles obtained for Staphylococcus hominis using forceps based rapid evaporative ionization mass spectrometry with IPA; and

Fig. 31A shows forceps based rapid evaporative ionization mass spectrometry spectra profiles for Pseudomonas aeruginosa with IPA and Fig. 31B shows forceps based rapid evaporative ionization mass spectrometry spectra profiles for Pseudomonas aeruginosa without IPA.


DETAILED DESCRIPTION



[0076] Various embodiments will now be described in more detail below which in general relate to an ion imager having an ambient ionization ion source device.

[0077] A plurality of different locations on a sample are automatically sampled using the device, and mass spectral data and/or ion mobility data corresponding to each location is obtained. The obtained mass spectral data and/or ion mobility data is then used to construct, train or improve a sample classification model.

Ambient ionization ion sources



[0078] According to various embodiments a first device is arranged and adapted to generate an aerosol, smoke or vapour from a sample (e.g., in vivo tissue). The device may comprise an ambient ionization ion source which is characterized by the ability to generate analyte aerosol, smoke or vapour from a native or unmodified sample. For example, other types of ionization ion sources such as Matrix Assisted Laser Desorption Ionization ("MALDI") ion sources require a matrix or reagent to be added to the sample prior to ionization.

[0079] It will be apparent that the requirement to add a matrix or a reagent to a sample prevents the ability to perform in vivo analysis of tissue and also, more generally, prevents the ability to provide a rapid simple analysis of target material.

[0080] In contrast, therefore, ambient ionization techniques are particularly advantageous since firstly they do not require the addition of a matrix or a reagent (and hence are suitable for the analysis of in vivo tissue) and since secondly they enable a rapid simple analysis of target material to be performed.

[0081] A number of different ambient ionization techniques are known and are intended to fall within the scope of the present invention. As a matter of historical record, Desorption Electrospray Ionization ("DESI") was the first ambient ionization technique to be developed and was disclosed in 2004. Since 2004, a number of other ambient ionization techniques have been developed. These ambient ionization techniques differ in their precise ionization method but they share the same general capability of generating gas-phase ions directly from native (i.e. untreated or unmodified) samples. A particular advantage of the various ambient ionization techniques which are intended to fall within the scope of the present invention is that the various ambient ionization techniques do not require any prior sample preparation. As a result, the various ambient ionization techniques enable both in vivo tissue and ex vivo tissue samples to be analyzed without necessitating the time and expense of adding a matrix or reagent to the tissue sample or other target material.

[0082] A list of ambient ionization techniques are given in the following Table 1:
Table 1: A list of ambient ionization techniques.
AcronymIonisation technique
DESI Desorption electrospray ionization
DeSSI Desorption sonic spray ionization
DAPPI Desorption atmospheric pressure photoionization
EASI Easy ambient sonic-spray ionization
JeDI Jet desorption electrospray ionization
TM-DESI Transmission mode desorption electrospray ionization
LMJ-SSP Liquid microjunction-surface sampling probe
DICE Desorption ionization by charge exchange
Nano-DESI Nanospray desorption electrospray ionization
EADESI Electrode-assisted desorption electrospray ionization
APTDCI Atmospheric pressure thermal desorption chemical ionization
V-EASI Venturi easy ambient sonic-spray ionization
AFAI Air flow-assisted ionization
LESA Liquid extraction surface analysis
PTC-ESI Pipette tip column electrospray ionization
AFADESI Air flow-assisted desorption electrospray ionization
DEFFI Desorption electro-flow focusing ionization
ESTASI Electrostatic spray ionization
PASIT Plasma-based ambient sampling ionization transmission
DAPCI Desorption atmospheric pressure chemical ionization
DART Direct analysis in real time
ASAP Atmospheric pressure solid analysis probe
APTDI Atmospheric pressure thermal desorption ionization
PADI Plasma assisted desorption ionization
DBDI Dielectric barrier discharge ionization
FAPA Flowing atmospheric pressure afterglow
HAPGDI Helium atmospheric pressure glow discharge ionization
APGDDI Atmospheric pressure glow discharge desorption ionization
LTP Low temperature plasma
LS-APGD Liquid sampling-atmospheric pressure glow discharge
MIPDI Microwave induced plasma desorption ionization
MFGDP Microfabricated glow discharge plasma
RoPPI Robotic plasma probe ionization
PLASI Plasma spray ionization
MALDESI Matrix assisted laser desorption electrospray ionization
ELDI Electrospray laser desorption ionization
LDTD Laser diode thermal desorption
LAESI Laser ablation electrospray ionization
CALDI Charge assisted laser desorption ionization
LA-FAPA Laser ablation flowing atmospheric pressure afterglow
LADESI Laser assisted desorption electrospray ionization
LDESI Laser desorption electrospray ionization
LEMS Laser electrospray mass spectrometry
LSI Laser spray ionization
IR-LAMICI Infrared laser ablation metastable induced chemical ionization
LDSPI Laser desorption spray post-ionization
PAMLDI Plasma assisted multiwavelength laser desorption ionization
HALDI High voltage-assisted laser desorption ionization
PALDI Plasma assisted laser desorption ionization
ESSI Extractive electrospray ionization
PESI Probe electrospray ionization
ND-ESSI Neutral desorption extractive electrospray ionization
PS Paper spray
DIP-APCI Direct inlet probe-atmospheric pressure chemical ionization
TS Touch spray
Wooden-tip Wooden-tip electrospray
CBS-SPME Coated blade spray solid phase microextraction
TSI Tissue spray ionization
RADIO Radiofrequency acoustic desorption ionization
LIAD-ESI Laser induced acoustic desorption electrospray ionization
SAWN Surface acoustic wave nebulization
UASI Ultrasonication-assisted spray ionization
SPA-nanoESI Solid probe assisted nanoelectrospray ionization
PAUSI Paper assisted ultrasonic spray ionization
DPESI Direct probe electrospray ionization
ESA-Py Electrospray assisted pyrolysis ionization
APPIS Ambient pressure pyroelectric ion source
RASTIR Remote analyte sampling transport and ionization relay
SACI Surface activated chemical ionization
DEMI Desorption electrospray metastable-induced ionization
REIMS Rapid evaporative ionization mass spectrometry
SPAM Single particle aerosol mass spectrometry
TDAMS Thermal desorption-based ambient mass spectrometry
MAII Matrix assisted inlet ionization
SAII Solvent assisted inlet ionization
SwiFERR Switched ferroelectric plasma ionizer
LPTD Leidenfrost phenomenon assisted thermal desorption


[0083] According to an embodiment the ambient ionisation ion source comprises a laser ionisation ion source. According to an embodiment the laser ionisation ion source may comprise a mid-IR laser ablation ion source. For example, there are several lasers which emit radiation close to or at 2.94 µm which corresponds with the peak in the water absorption spectrum. According to various embodiments the ambient ionisation ion source may comprise a laser ablation ion source having a wavelength close to 2.94 µm on the basis of the high absorption coefficient of water at 2.94 µm. According to an embodiment the laser ablation ion source may comprise a Er:YAG laser which emits radiation at 2.94 µm.

[0084] Other embodiments are contemplated wherein a mid-infrared optical parametric oscillator ("OPO") may be used to produce a laser ablation ion source having a longer wavelength than 2.94 µm. For example, an Er:YAG pumped ZGP-OPO may be used to produce laser radiation having a wavelength of e.g. 6.1 µm, 6.45 µm or 6.73 µm. In some situations it may be advantageous to use a laser ablation ion source having a shorter or longer wavelength than 2.94 µm since only the surface layers will be ablated and less thermal damage may result. According to an embodiment a Co:MgF2 laser may be used as a laser ablation ion source wherein the laser may be tuned from 1.75-2.5 µm. According to another embodiment an optical parametric oscillator ("OPO") system pumped by a Nd:YAG laser may be used to produce a laser ablation ion source having a wavelength between 2.9-3.1 µm. According to another embodiment a CO2 laser having a wavelength of 10.6 µm may be used to generate the aerosol, smoke or vapour.

Rapid evaporative ionization mass spectrometry ("REIMS")



[0085] Fig. 1 illustrates a method of rapid evaporative ionization mass spectrometry ("REIMS") wherein bipolar forceps 1 may be brought into contact with in vivo tissue 2 of a patient 3. In the example shown in Fig. 1, the bipolar forceps 1 may be brought into contact with brain tissue 2 of a patient 3 during the course of a surgical operation on the patient's brain. An RF voltage from an RF voltage generator 4 may be applied to the bipolar forceps 1 which causes localised Joule or diathermy heating of the tissue 2. As a result, an aerosol or surgical plume 5 is generated. The aerosol or surgical plume 5 may then be captured or otherwise aspirated through an irrigation port of the bipolar forceps 1. The irrigation port of the bipolar forceps 1 is therefore reutilised as an aspiration port. The aerosol or surgical plume 5 may then be passed from the irrigation (aspiration) port of the bipolar forceps 1 to tubing 6 (e.g. 1/8" or 3.2 mm diameter Teflon (RTM) tubing). The tubing 6 is arranged to transfer the aerosol or surgical plume 5 to an atmospheric pressure interface 7 of a mass spectrometer and/or ion mobility spectrometer 8.

[0086] According to various embodiments a matrix comprising an organic solvent such as isopropanol (IPA) may be added to the aerosol or surgical plume 5 at the atmospheric pressure interface 7. The mixture of aerosol 3 and organic solvent may then be arranged to impact upon a collision surface within a vacuum chamber of the mass spectrometer and/or ion mobility spectrometer 8. According to one embodiment the collision surface may be heated. The aerosol is caused to ionize upon impacting the collision surface resulting in the generation of analyte ions. The ionization efficiency of generating the analyte ions may be improved by the addition of the organic solvent. However, the addition of an organic solvent is not essential.

[0087] Analyte ions which are generated by causing the aerosol, smoke or vapour 5 to impact upon the collision surface are then passed through subsequent stages of the mass spectrometer and/or ion mobility spectrometer and are subjected to mass analysis and/or ion mobility analysis in a mass analyser and/or ion mobility analyser. The mass analyser may, for example, comprise a quadrupole mass analyser or a Time of Flight mass analyser.

Sample treatment



[0088] For the analysis of human samples, ethical approval was obtained from the National Healthcare Service Research Ethics Committee (Study ID 11/LO/1686).

[0089] Fig. 2 shows a rapid evaporative ionization mass spectrometry imaging platform (i.e. an ion imager) located above a tissue sample 20 to be imaged. The rapid evaporative ionization mass spectrometry imaging platform includes a first device which may comprise a sampling needle, electrode, tip or probe 21 that is brought into contact with a sample 20 to generate gaseous or aerosolised analyte material by the rapid evaporative ionization process (i.e. a first device arranged and adapted to generate aerosol, smoke or vapour from the sample). Power may be provided to the sampling probe 21 by a high voltage power supply 23, in conjunction with a function generator 23a. The evaporated gaseous or aerosolised analyte material may be captured by suction tubing 22 and transported (i.e. aspirated) through tubing 24 towards a mass spectrometer and/or ion mobility spectrometer 28 for analysis.

[0090] The sampling probe 21 may be mounted onto a z-actuator and may be manipulated over the sample 20 in the x-y plane to automatically sample and generate analyte material at a plurality of different locations over the whole area of the sample 20. Correlating the position of the sampling needle 21 relative to the xyz stage 25 with the results of the mass spectrometric and/or ion mobility analysis allows ion imaging of the sample 20.

[0091] Thus a plurality of different locations on the sample 20 may be automatically sampled using the first device, which is arranged and adapted to generate aerosol, smoke or vapour from the sample. By obtaining mass spectral data and/or ion mobility data corresponding to each of the locations, an ion image, such as ion image 26, may be generated.

[0092] Alternatively or additionally, the obtained mass spectral data and/or ion mobility spectrometer may be used to construct, train or improve a sample classification model. For example, the sample classification model represented by principle components analysis (PCA) loadings 27.

[0093] Fig. 3 shows a workflow of combined Desorption Electrospray Ionization ("DESI") and rapid evaporative ionization mass spectrometry imaging platform analysis for co-registration of histological features between optical image, Desorption Electrospray Ionization ("DESI") and rapid evaporative ionization mass spectrometry data. Fresh human liver metastasis samples were obtained from surgical resection specimens and immediately frozen to -80°C, at workflow stage 30. At stages 31a and 31b, the tissue samples were cryosectioned (Thermo Microm HM550 Cryostat, Thermo Fisher Scientific (RTM), Germany) to 10 µm thickness and thaw mounted onto glass slides for Desorption Electrospray lonisation ("DESI") analysis, as illustrated at workflow stage 32. The remaining bulk tissue was used for rapid evaporative ionization mass spectrometry analysis, as illustrated at stage 33.

[0094] Desorption Electrospray Ionization ("DESI") imaging analysis on the glass slide mounted tissue sample was carried out using an in-house built Desorption Electrospray Ionization ("DESI") stage at stage 34, to generate a Desorption Electrospray Ionization ("DESI") ion image illustrated at stage 36. At workflow stage 35, rapid evaporative ionization mass spectrometry imaging analysis on the bulk tissue sample was performed using a modified Prosolia (RTM) flowprobe stage (Prosolia (RTM), USA), to generate rapid evaporative ionization mass spectrometry ion images, for example ion images illustrated at 39a and 39b.

[0095] Desorption Electrospray Ionization ("DESI") analysis of tissues was carried out using a mass spectrometer operated in negative ion mode.

[0096] The Desorption Electrospray Ionization ("DESI") imaging pixel size was set to 100 µm, the electrospray solvent was methanol:water (95:5 vol/vol) at a solvent flow rate of 1.5 µL/min and zero-grade nitrogen nebulizing gas at a pressure of 4 bar was used. Following Desorption Electrospray Ionization ("DESI") analysis, at stage 37, tissue sections were stained with H&E (haematoxylin and eosin) and digitally scanned (Nano-Zoomer 2.0-HT, Hamamatsu (RTM), Japan) to create optical images at stage 38 for comparison with the ambient ionization mass spectral (Desorption Electrospray Ionization ("DESI") and rapid evaporative ionization mass spectrometry) images.

[0097] A line scan mode (cutting mode of operation) rapid evaporative ionization mass spectrometry analysis of one liver metastasis sample was performed on a mass spectrometer and a spot sampling (pointing mode of operation) analysis of another liver metastasis sample and a microorganism culture were performed on a Waters Xevo G2-S Q-TOF instrument (RTM) (Waters Micromass (RTM), U.K.) in negative ion mode.

[0098] The Waters Xevo G2-S (RTM) mass spectrometer was equipped with a modified atmospheric interface 40 combining an orthogonal Venturi-pump for aerosol transfer and a heated capillary inlet as shown in Fig. 4. The heated capillary inlet comprises a cylindrical collision surface 44 mounted within a ceramic holder 43 and heated via a sheathed 42 conductive coil 41. The use of such a heated coil interface may provide improved sensitivity and robustness against contamination.

[0099] Thus, at least some aerosol, smoke or vapour generated by a first device operating in a cutting or pointing mode of operation may be caused to impact upon the heated collision surface located within the vacuum chamber of a mass spectrometer and/or ion mobility spectrometer, so as to generate analyte ions.

[0100] Rapid evaporative ionization mass spectrometry imaging analysis of liver metastasis was carried out in a (first) cutting mode at 1 bar Venturi gas pressure and about 4 kV p-p amplitude at about 50 kHz alternating current frequency (AC). A blade-shaped electrosurgical tip (sampling probe) was used, about 500 µm pixel size, 1 mm/s cutting speed and 1 mm cutting depth.

[0101] Analysis of liver metastasis in a (second) pointing mode was carried out at about 0.25 bar Venturi gas pressure, 2 kV amplitude at about 50 kHz AC and using a wire-shaped electrosurgical tip at about 750 µm pixel size, 0.1 s time remaining inside the sample and a pointing depth of about 1 mm.

[0102] Aerosol was transferred (i.e. aspirated) using a 1/8" OD, 2 mm ID PTFE tubing. Since the used power settings were sufficiently high such as potentially to cause severe injury, the instrumental setup was handled with high caution and insulating gloves were worn.

[0103] Parameter optimization of the rapid evaporative ionization mass spectrometry imaging platform was carried out using porcine liver samples. For comparison of mass spectral patterns between rapid evaporative ionization mass spectrometry imaging and iKnife technology, porcine liver, porcine kidney cortex, lamb liver and chicken skeletal muscle were analysed using an electrosurgical handpiece (Meyer-Haake GmbH (RTM), Germany) with incorporated PTFE tubing (1/8" OD, 2mm ID) which was connected to the Venturi pump. Liver, kidney and muscle were food grade and purchased as such. The iKnife technology was operated in a cutting mode at 40 W and 1 bar gas pressure in combination with a Valleylab SurgiStat II (RTM) power-controlled electrosurgical generator (Covidien, Ireland).

Data processing



[0104] Raw spectral profiles were loaded into a MATLAB (RTM) environment (Version R2014a, Mathworks, USA) for pre-processing, MS-image visualization and pattern recognition analysis. All mass spectra were linearly interpolated to a common interval of 0.1 Da and individually normalized to the total ion count ("TIC") of each mass spectrum. The data was used for univariate comparison of intensity levels across liver tissue types and ionization techniques and for bacterial MS-image visualization of single ions. Peak annotation for liver metastasis samples was based on m/z accuracy obtained from the unprocessed raw files, while bacterial peak annotation was based on mass accuracy and on tandem-MS spectra obtained using bipolar forceps.

[0105] Multivariate MS-image visualization was performed on mass spectra additionally binned to 1 Da intervals in the mass range of m/z 600-1000 Da for biological tissue and m/z 400-2000 for bacteria. For multivariate image visualization, MS-images and optical images were co-registered to define regions of interest ("ROIs") for building a supervised training model (i.e. a sample classification model). Defined ROIs (classes) were healthy and cancerous tissue for the liver samples and one region for each bacterium plus agar, resulting overall in two classes for liver samples and four classes for bacterial samples.

[0106] The training model was used to classify each pixel of the same sample and colour code the obtained score-values into red-green-blue colour scale. This supervised strategy for image visualization is based on an algorithm that combines recursive maximum margin criterion ("RMMC") with linear discriminant analysis ("LDA"). For unsupervised analysis, principal component analysis ("PCA") was performed on the mass spectra defined by the regions of interest.

[0107] Concordance correlation coefficients were used to measure the agreement between rapid evaporative ionization mass spectrometry imaging platform ("RIP") mass spectra and iKnife technology mass spectra. This quantitative measure is defined as:

wherein ρc is the concordance correlation coefficient, ρ is Pearson's correlation coefficient and σRIP/iKnife is the standard deviation of the mean intensity values of µRIP/iKnife.

[0108] A low concordance correlation coefficient close to the value of zero indicates low agreement while a value close to the value of one suggests high similarity between spectral profiles.

[0109] Boxplots show the median at the central mark within the box with 25th and 75th percentiles at the edges of the box. The upper and lower whiskers account for approximately 2.7 standard deviations (99.3% data coverage). Mass spectra were standardized to 100% intensity scale before their data was visualized with boxplots.

Analysing sample spectra



[0110] A list of analysis techniques which are intended to fall within the scope of the present invention are given in the following Table 2:
Table 2: A list of analysis techniques.
Analysis Techniques
Univariate Analysis
Multivariate Analysis
Principal Component Analysis (PCA)
Linear Discriminant Analysis (LDA)
Maximum Margin Criteria (MMC)
Library Based Analysis
Soft Independent Modelling Of Class Analogy (SIMCA)
Factor Analysis (FA)
Recursive Partitioning (Decision Trees)
Random Forests
Independent Component Analysis (ICA)
Partial Least Squares Discriminant Analysis (PLS-DA)
Orthogonal (Partial Least Squares) Projections To Latent Structures (OPLS)
OPLS Discriminant Analysis (OPLS-DA)
Support Vector Machines (SVM)
(Artificial) Neural Networks
Multilayer Perceptron
Radial Basis Function (RBF) Networks
Bayesian Analysis
Cluster Analysis
Kernelized Methods
Subspace Discriminant Analysis
K-Nearest Neighbours (KNN)
Quadratic Discriminant Analysis (QDA)
Probabilistic Principal Component Analysis (PPCA)
Non negative matrix factorisation
K-means factorisation
Fuzzy c-means factorisation
Discriminant Analysis (DA)


[0111] Combinations of the foregoing analysis approaches can also be used, such as PCA-LDA, PCA-MMC, PLS-LDA, etc.

[0112] Analysing the sample spectra can comprise unsupervised analysis for dimensionality reduction followed by supervised analysis for classification.

[0113] By way of example, a number of different analysis techniques will now be described in more detail.

Multivariate analysis - developing a model for classification



[0114] According to various embodiments, obtained mass spectral data and/or ion mobility data is used to construct, train or improve a sample classification model. By way of example, a method of building a classification model using multivariate analysis of plural reference sample spectra will now be described.

[0115] Fig. 5 shows a method 500 of building a classification model using multivariate analysis according to an embodiment. In this example, the method comprises a step 502 of obtaining plural sets of intensity values for reference sample spectra. The method then comprises a step 504 of unsupervised principal component analysis (PCA) followed by a step 506 of supervised linear discriminant analysis (LDA). This approach may be referred to herein as PCA-LDA. Other multivariate analysis approaches may be used, such as PCA-MMC. The PCA-LDA model is then output, for example to storage, in step 508.

[0116] The multivariate analysis such as this can provide a classification model that allows an aerosol, smoke or vapour sample to be classified using one or more sample spectra obtained from the aerosol, smoke or vapour sample. The multivariate analysis will now be described in more detail with reference to a simple example.

[0117] Fig. 6 shows a set of reference sample spectra obtained from two classes of known reference samples. The classes may be any one or more of the classes of target described herein. However, for simplicity, in this example the two classes will be referred as a lefthand class and a right-hand class.

[0118] Each of the reference sample spectra has been pre-processed in order to derive a set of three reference peak-intensity values for respective mass to charge ratios in that reference sample spectrum. Although only three reference peak-intensity values are shown, it will be appreciated that many more reference peak-intensity values (e.g., ∼ 100 reference peak-intensity values) may be derived for a corresponding number of mass to charge ratios in each of the reference sample spectra. In other embodiments, the reference peak-intensity values may correspond to: masses; mass to charge ratios; ion mobilities (drift times); and/or operational parameters.

[0119] Fig. 7 shows a multivariate space having three dimensions defined by intensity axes. Each of the dimensions or intensity axes corresponds to the peak-intensity at a particular mass to charge ratio. Again, it will be appreciated that there may be many more dimensions or intensity axes (e.g., - 100 dimensions or intensity axes) in the multivariate space. The multivariate space comprises plural reference points, with each reference point corresponding to a reference sample spectrum, i.e., the peak-intensity values of each reference sample spectrum provide the co-ordinates for the reference points in the multivariate space.

[0120] The set of reference sample spectra may be represented by a reference matrix D having rows associated with respective reference sample spectra, columns associated with respective mass to charge ratios, and the elements of the matrix being the peak-intensity values for the respective mass to charge ratios of the respective reference sample spectra.

[0121] In many cases, the large number of dimensions in the multivariate space and matrix D can make it difficult to group the reference sample spectra into classes. PCA may accordingly be carried out on the matrix D in order to calculate a PCA model that defines a PCA space having a reduced number of one or more dimensions defined by principal component axes. The principal components may be selected to be those that comprise or "explain" the largest variance in the matrix D and that cumulatively explain a threshold amount of the variance in the matrix D.

[0122] Fig. 8 shows how the cumulative variance may increase as a function of the number n of principal components in the PCA model. The threshold amount of the variance may be selected as desired.

[0123] The PCA model may be calculated from the matrix D using a non-linear iterative partial least squares (NIPALS) algorithm or singular value decomposition, the details of which are known to the skilled person and so will not be described herein in detail. Other methods of calculating the PCA model may be used.

[0124] The resultant PCA model may be defined by a PCA scores matrix S and a PCA loadings matrix L. The PCA may also produce an error matrix E, which contains the variance not explained by the PCA model. The relationship between D, S, L and E may be:



[0125] Fig. 9 shows the resultant PCA space for the reference sample spectra of Figs. 6 and 7. In this example, the PCA model has two principal components PC0 and PC1 and the PCA space therefore has two dimensions defined by two principal component axes. However, a lesser or greater number of principal components may be included in the PCA model as desired. It is generally desired that the number of principal components is at least one less than the number of dimensions in the multivariate space.
The PCA space comprises plural transformed reference points or PCA scores, with each transformed reference point or PCA score corresponding to a reference sample spectrum of Fig. 6 and therefore to a reference point of Fig. 7.

[0126] As is shown in Fig. 9, the reduced dimensionality of the PCA space makes it easier to group the reference sample spectra into the two classes. Any outliers may also be identified and removed from the classification model at this stage.

[0127] Further supervised multivariate analysis, such as multi-class LDA or maximum margin criteria (MMC), in the PCA space may then be performed so as to define classes and, optionally, further reduce the dimensionality.

[0128] As will be appreciated by the skilled person, multi-class LDA seeks to maximise the ratio of the variance between classes to the variance within classes (i.e., so as to give the largest possible distance between the most compact classes possible). The details of LDA are known to the skilled person and so will not be described herein in detail. The resultant PCA-LDA model may be defined by a transformation matrix U, which may be derived from the PCA scores matrix S and class assignments for each of the transformed spectra contained therein by solving a generalised eigenvalue problem.

[0129] The transformation of the scores S from the original PCA space into the new LDA space may then be given by:

where the matrix Z contains the scores transformed into the LDA space.

[0130] Fig. 10 shows a PCA-LDA space having a single dimension or axis, wherein the LDA is performed in the PCA space of Fig. 9. As is shown in Fig. 10, the LDA space comprises plural further transformed reference points or PCA-LDA scores, with each further transformed reference point corresponding to a transformed reference point or PCA score of Fig. 9.

[0131] In this example, the further reduced dimensionality of the PCA-LDA space makes it even easier to group the reference sample spectra into the two classes. Each class in the PCA-LDA model may be defined by its transformed class average and covariance matrix or one or more hyperplanes (including points, lines, planes or higher order hyperplanes) or hypersurfaces or Voronoi cells in the PCA-LDA space.

[0132] The PCA loadings matrix L, the LDA matrix U and transformed class averages and covariance matrices or hyperplanes or hypersurfaces or Voronoi cells may be output to a database for later use in classifying an aerosol, smoke or vapour sample.

[0133] The transformed covariance matrix in the LDA space V'g for class g may be given by:

where Vg are the class covariance matrices in the PCA space.

[0134] The transformed class average position zg for class g may be given by:

where sg is the class average position in the PCA space.

Multivariate analysis - using a model for classification



[0135] According to various embodiments, a sample classification model which was previously constructed, trained or improved according to a method described herein is used in order to classify a sample at a location. By way of example, a method of using a classification model to classify an aerosol, smoke or vapour sample will now be described.

[0136] Fig. 11 shows a method 1100 of using a classification model according to an embodiment. In this example, the method comprises a step 1102 of obtaining a set of intensity values for a sample spectrum. The method then comprises a step 1104 of projecting the set of intensity values for the sample spectrum into PCA-LDA model space. Other classification model spaces may be used, such as PCA-MMC. The sample spectrum is then classified at step 1106 based on the project position and the classification is then output in step 1108.

[0137] Classification of an aerosol, smoke or vapour sample will now be described in more detail with reference to the simple PCA-LDA model described above.

[0138] Fig. 12 shows a sample spectrum obtained from an unknown aerosol, smoke or vapour sample. The sample spectrum has been pre-processed in order to derive a set of three sample peak-intensity values for respective mass to charge ratios. As mentioned above, although only three sample peak-intensity values are shown, it will be appreciated that many more sample peak-intensity values (e.g., - 100 sample peak-intensity values) may be derived at many more corresponding mass to charge ratios for the sample spectrum. Also, as mentioned above, in other embodiments, the sample peak-intensity values may correspond to: masses; mass to charge ratios; ion mobilities (drift times); and/or operational parameters.

[0139] The sample spectrum may be represented by a sample vector dx, with the elements of the vector being the peak-intensity values for the respective mass to charge ratios. A transformed PCA vector sX for the sample spectrum can be obtained as follows:



[0140] Then, a transformed PCA-LDA vector zX for the sample spectrum can be obtained as follows:



[0141] Fig. 13 again shows the PCA-LDA space of Fig. 10. However, the PCA-LDA space of Fig. 13 further comprises the projected sample point, corresponding to the transformed PCA-LDA vector zx, derived from the peak intensity values of the sample spectrum of Fig. 12.

[0142] In this example, the projected sample point is to one side of a hyperplane between the classes that relates to the right-hand class, and so the aerosol, smoke or vapour sample may be classified as belonging to the right-hand class.

[0143] Alternatively, the Mahalanobis distance from the class centres in the LDA space may be used, where the Mahalanobis distance of the point zx from the centre of class g may be given by the square root of:

and the data vector dx may be assigned to the class for which this distance is smallest.

[0144] In addition, treating each class as a multivariate Gaussian, a probability of membership of the data vector to each class may be calculated.

Library based analysis - developing a library for classification



[0145] By way of example, a method of building a classification library using plural input reference sample spectra will now be described.

[0146] Fig. 14 shows a method 1400 of building a classification library. In this example, the method comprises a step 1402 of obtaining plural input reference sample spectra and a step 1404 of deriving metadata from the plural input reference sample spectra for each class of sample. The method then comprises a step 1404 of storing the metadata for each class of sample as a separate library entry. The classification library is then output, for example to electronic storage, in step 1406.

[0147] A classification library such as this allows an aerosol, smoke or vapour sample to be classified using one or more sample spectra obtained from the aerosol, smoke or vapour sample. The library based analysis will now be described in more detail with reference to an example.

[0148] In this example, each entry in the classification library is created from plural pre-processed reference sample spectra that are representative of a class. In this example, the reference sample spectra for a class are pre-processed according to the following procedure:
First, a re-binning process is performed. In this embodiment, the data are resampled onto a logarithmic grid with abscissae:

where Nchan is a selected value and |x| denotes the nearest integer below x. In one example, Nchan is 212 or 4096.

[0149] Then, a background subtraction process is performed. In this embodiment, a cubic spline with k knots is then constructed such that p% of the data between each pair of knots lies below the curve. This curve is then subtracted from the data. In one example, k is 32. In one example, p is 5.

[0150] A constant value corresponding to the q% quantile of the intensity subtracted data is then subtracted from each intensity. Positive and negative values are retained. In one example, q is 45.

[0151] Then, a normalisation process is performed. In this embodiment, the data are normalised to have mean yi. In one example, yi = 1.

[0152] An entry in the library then consists of metadata in the form of a median spectrum value µi and a deviation value Di for each of the Nchan points in the spectrum.

[0153] The likelihood for the i'th channel is given by:

where 1/2 ≤ C < ∞ and where Γ(C) is the gamma function.

[0154] The above equation is a generalised Cauchy distribution which reduces to a standard Cauchy distribution for C = 1 and becomes a Gaussian (normal) distribution as C → ∞. The parameter Di controls the width of the distribution (in the Gaussian limit Di = σi is simply the standard deviation) while the global value C controls the size of the tails.

[0155] In one example, C is 3/2, which lies between Cauchy and Gaussian, so that the likelihood becomes:

For each library entry, the parameters µi are set to the median of the list of values in the i'th channel of the input reference sample spectra while the deviation Di is taken to be the interquartile range of these values divided by √2. This choice can ensure that the likelihood for the i'th channel has the same interquartile range as the input data, with the use of quantiles providing some protection against outlying data.

Library based analysis - using a library for classification



[0156] By way of example, a method of using a classification library to classify an aerosol, smoke or vapour sample will now be described.

[0157] Fig. 15 shows a method 1500 of using a classification library. In this example, the method comprises a step 1502 of obtaining a set of plural sample spectra. The method then comprises a step 1504 of calculating a probability or classification score for the set of plural sample spectra for each class of sample using metadata for the class entry in the classification library. The sample spectra are then classified at step 1506 and the classification is then output in step 1508.

[0158] Classification of an aerosol, smoke or vapour sample will now be described in more detail with reference to the classification library described above.

[0159] In this example, an unknown sample spectrum y is the median spectrum of a set of plural sample spectra. Taking the median spectrum y can protect against outlying data on a channel by channel basis.

[0160] The likelihood Ls for the input data given the library entry s is then given by:

where µi and Di are, respectively, the library median values and deviation values for channel i. The likelihoods Ls may be calculated as log likelihoods for numerical safety.

[0161] The likelihoods Ls are then normalised over all candidate classes's' to give probabilities, assuming a uniform prior probability over the classes. The resulting probability for the class is given by:



[0162] The exponent (1/F) can soften the probabilities which may otherwise be too definitive. In one example, F = 100. These probabilities may be expressed as percentages, e.g., in a user interface.

[0163] Alternatively, RMS classification scores Rs may be calculated using the same median sample values and derivation values from the library:



[0164] Again, the scores Rs are normalised over all candidate classes 's'.

[0165] The aerosol, smoke or vapour sample may then be classified as belonging to the class having the highest probability and/or highest RMS classification score.

Rapid evaporative ionization mass spectrometry imaging platform



[0166] Fig. 16 schematically illustrates and Figs. 17A-C show a rapid evaporative ionization mass spectrometry imaging platform (i.e. an ion imager), which includes three major functional elements that all influence the quality of mass spectra. The imaging platform may include a power generator, a xyz-stage with a sampling probe (i.e. a first device arranged and adapted to generate aerosol, smoke or vapour from a sample) and a mass spectrometer and/or ion mobility spectrometer.

[0167] The mass spectral data and/or ion mobility data obtained using the rapid evaporative ionization mass spectrometry (or other ambient ionisation) imaging platform may be used to construct, train or improve a sample classification model (e.g., as described above).

[0168] The power supply setup used for the platform may comprise a Tektronix (RTM) AFG 3022 arbitrary function generator (Tektronix (RTM), USA), a Tektronix (RTM) DPO 3014 Oscilloscope and a Trek 10/40A High Voltage Amplifier (Trek (RTM), USA).

[0169] The arbitrary function generator was used to generate sinus waveforms with amplitudes between 1 V and 6 V at frequencies in the range of 10 to 60 kHz. The high voltage power amplifier multiplied the voltage by a factor of 1000 and supplied the connected sampling probe with the electric current. The oscilloscope provided feedback to ensure correct working parameters.

[0170] The xyz-stage may comprise a modified Prosolia (RTM) 2D Desorption Electrospray Ionization ("DESI") stage 131 (as shown in Fig. 17C) and may include a Flowprobe (RTM) upgrade (Prosolia (RTM), USA) with a high precision z-axis actuator 132. The sampling probe 21 is mounted onto the actuator 132 via two mounting points 133 and is connected to the power generator setup 23 as well as a MS inlet capillary through tubing 24 (as shown in Fig. 17A).

[0171] A laser height sensor 134 may be used to measure the distance or height between an electrosurgical tip of the sampling probe 21 (or more generally the first device) and the sample surface, and can ensure an equal penetration depth of the tip into the sample which is useful for uneven sample surfaces. The laser height sensor 134 may comprise a camera. The electrosurgical tip of the sampling probe 21 may be exchanged for other materials or shapes depending on the field of application. In case of high precision sampling, a small diameter wire may be used, whereas a large surface tip is suitable to maximize mass spectrometric and/or ion mobility signal intensity. A variety of possible alternatively shaped sample probes are shown in Fig. 17B. The electrosurgical tip may be surrounded by suction tubing 22 which is connected to a Venturi air jet pump.

[0172] An optical fibre in conjunction with a laser source may be used to generate aerosol, smoke or vapour from a target (e.g. tissue sample).

[0173] The imaging platform is capable of at least two sampling modes; namely a cutting mode of operation as illustrated in Fig. 18A and a pointing mode of operation as illustrated in Fig. 18B. In a cutting mode of operation, line scans are performed and the electrosurgical tip of the sampling probe 21 may be kept at a constant z-value, i.e. a constant height above the sample, while the x and y values can change in a way such that a macroscopic cut is made in a right to left trajectory through the tissue sample, with each subsequent cut being made at an increment further in the y direction. In this approach, the electrosurgical tip is in substantially continuous contact with the sample and therefore continuously produces aerosol.

[0174] The speed of x-movement influences the width of the region of tissue disruption and the amount of aerosol produced (as illustrated in Fig. 19A). If the step size in the y direction is smaller than the burning-valley-width, then a complete surface layer will be evaporated (as illustrated in Fig. 19C).

[0175] In a pointing mode of operation, the sampling probe 21 can penetrate the sample for a given depth and time. Both factors influence the amount of evaporated aerosol and burn-crater size as is apparent from Fig. 19B.

[0176] In terms of imaging performance, the time of contact between the electrosurgical tip and the sample can influence the achievable spatial resolution which is limited by the width of tissue disruption. As ion current is also a function of cutting speed, there is (like in the case of all other MSI methods) a trade-off between spatial resolution, signal intensity and sampling time. In a cutting mode of operation, the speed of imaging depends on a user defined cutting speed which is usually the already mentioned compromise between mass spectrometer and/or ion mobility spectrometer sampling time and desired spatial resolution.

[0177] In the case of a pointing mode of operation, the time necessary to move from one sampling spot or location to the next may be determined by the maximum movement speed of the xyz-stage and the time the sampling probe tip remains inside the sample. An exemplary cutting speed is about 1 mm/s, and the time necessary to record one pixel in a pointing mode of operation may be about 3 s, for example. Using these parameters, imaging of a 2x2 cm sample with 2 mm spatial resolution will take an approximately equal amount of time of about 5 minutes for both pointing and cutting modes of operation (see Table 3 below). The additional time necessary to move the z-actuator in the pointing mode of operation becomes more significant as the pixel size becomes smaller. This leads to a five times higher amount of imaging time at 500 µm pixel size in a pointing mode of operation compared with a cutting mode of operation.

[0178] While cutting mode imaging at low resolutions evaporates the whole top sample layer, pointing mode in low resolution leaves the majority of tissue unaffected, allowing the same surface to be characterized at a later time.

[0179] In both cases, the user of a preferred rapid evaporative ionization mass spectrometry imaging platform (i.e. ion imager) should be aware of the heterogeneity within the sample, as cutting and pointing depth causes tissue evaporation from the bulk sample.
Table 3: Theoretical sampling time and resolution for 2x2 cm sample. Cutting mode sampling at 1 mm/s cutting speed and 25 s per row, which includes return time to a new row. Pointing mode sampling at 3 s per pixel.
  Pointing ModeCutting Mode
Pixel SizeNo. of PixelsTime / minNo. of RowsMS scan time / sTime / min
2 mm 100 5 10 2 4.2
1 mm 400 20 20 1 8.3
500 µm 1600 80 40 0.5 16.7
250 µm 6400 320 80 0.25 33.3


[0180] The transfer (i.e. aspiration) of aerosol to the mass spectrometer and/or ion mobility spectrometer may be carried out using a Venturi air jet pump mounted to an atmospheric interface of a mass spectrometer and/or ion mobility spectrometer. The aerosol trajectory may be perpendicular to the MS-inlet capillary. As a result, larger particles may be excluded by momentum separation thereby avoiding clogging and contamination of the mass spectrometer and/or ion mobility spectrometer. Excess aerosol may be captured by a surgical smoke trap device.

Frequency and voltage dependencies



[0181] The imaging platform (i.e. ion imager) can enable automated high-throughput collection of reference mass spectra and/or ion mobility data in order to aid real-time classification in MS-guided electrosurgery (iKnife technology) applications. For example, according to an embodiment, the classification algorithm (i.e. sample classification model) may compare mass spectral and/or ion mobility patterns of spectra created during surgery with mass spectra obtained ex vivo, in vivo or in vitro. Accordingly, it is important that the rapid evaporative ionization mass spectrometry imaging platform provides similar ionization conditions as will be used in surgery.

[0182] Thus, according to this embodiment, a plurality of different locations of a sample are sampled using a first device arranged and adapted to generate aerosol, smoke or vapour from the sample to obtain mass spectral data and/or ion mobility data at each location. A sample classification model which was previously constructed, trained or improved according to a method of ion imaging as described herein is then used in order to classify the sample at each location.

[0183] Commercially available electrosurgical generators as used in operating theatres provide highly reproducible mass spectral patterns which are unique for different histological tissue types. The power supply setup used in conjunction with the imaging platform (as shown schematically illustrated in Fig. 16) may allow variation in the amplitude and/or frequency and/or waveform, while an oscilloscope may provide feedback ensuring correct working conditions. Depending on the application of the imaging platform, the experimental parameters can thus be changed in order to alter ionization conditions and to meet the requirements for recording reference mass spectra for intra-surgical tissue identification or bacterial classification purposes.

[0184] Rapid evaporative ionization mass spectrometry ionization mechanism is based on Joule-heating which is a thermal process wherein the heat created is proportional to the square of electric current and the impedance. As electric current density is also a function of cross sectional area, the contact surface area of the electrosurgical tip of the sampling probe 21 also has an impact on the heating process.

[0185] If an electric current is applied to a biological tissue then the intracellular temperature rises up to a point of vaporization where excess heat facilitates evaporation of particles and ions leading to the formation of surgical aerosol. The major ions created in this process are singly charged lipids being most abundant in the m/z 600-1000 mass range for eukaryotic tissue and additionally in the m/z 1100-1500 mass range in case of bacteria in form of e.g. lipid dimers or cardiolipins.

[0186] Depending on the thermal stability of the molecules, thermal degradation may occur as it was observed in the case of phosphatidyl-ethanolamine species which are partly ionized to both [M-NH4]- and [M-H]-, while other phospholipids species form [M-H]- ions. The density and frequency of the electric current can therefore have an important influence on the appearance of the mass spectrum.

[0187] Electrosurgical generators have an incorporated control loop providing constant power when cutting through tissue, even if the impedance is rapidly changing. This leads to gentle and reproducible cuts with minimized tissue heat exposure. Electrosurgical generators are not easily incorporated into an imaging set up due to a number of safety measures required when used in theatre, hence a simplified power supply was built. Since a p-p voltage amplitude-controlled RF power supply cannot follow the changing impedance of the sample, it was important to determine whether the simplified setup can provide spectra similar to those obtained when using proper electrosurgical equipment.

[0188] Optimization of the rapid evaporative ionization mass spectrometry imaging platform was carried out by finding the optimal frequency and voltage values to match the iKnife technology reference mass spectral pattern of porcine liver as shown in Figs. 20A-B and Figs. 21A-B. Concordance correlation coefficients ("CCC") between the rapid evaporative ionization mass spectrometry imaging and iKnife technology mass spectra were used as a quantitative measure to find the optimal spectral agreement.

[0189] In cutting mode, a factor influencing tissue heat exposure is cutting speed, which leads to high localized temperature for slow speeds and vice versa. Depending on the required ion current, the MS sampling time window needs to be sufficiently long, compromising either spatial resolution or cutting speeds. Therefore, prior to voltage and frequency optimization, a cutting speed should be chosen that satisfies requirements on ion yield and spatial resolution. Once a cutting speed is set, heat exposure can then be controlled by changing the voltage or frequency output of the power generator setup. The cutting speed may need further reiteration if the available range of voltages and frequencies is not sufficient for adequate heat production. An exemplary cutting speed of 1 mm/s was found to gently cut at high ion yields.

[0190] As shown in Fig. 20A, at a constant p-p voltage of 2 kV an increase in frequency leads to less thermal degradation and higher similarity to iKnife technology patterns. According to the oscilloscope readout, the power generator setup was not capable of maintaining a constant increase in power output above 50 kHz at a 2 kV amplitude, explaining the stable concordance correlation coefficient between about 40 kHz and about 60 kHz. At lower frequencies more in-depth heat dissipation was observed leading to wide burning valleys, carbonization and inconsistent mass spectral patterns with varying baseline noise levels. This was accompanied by strong soot particle production leading to contamination of the MS-inlet capillary, without contributing to the ion yield (see the total ion counts in Fig. 21A).

[0191] At higher frequencies (above about 40 kHz) visible soot particle production was negligible and no carbonization was observed. This led to mass spectral patterns very similar to those produced by electrosurgical equipment, as indicated by concordance correlation coefficients near 0.9. The highest and most consistent TIC was also found to be in that frequency window.

[0192] As shown in Fig. 20B, an increase in voltage at 40 kHz frequency resulted in similar phenomena as observed with decreasing frequency, such as carbonization and wide burning valleys, leading to high concordance correlation coefficients to be found at low voltages. However, once the voltage was set below about 2 kV, ion currents dramatically dropped (see the total ion counts in Fig. 21B). This led to an optimal parameter window between about 3-4 kV and about 40-50 kHz where concordance correlation coefficients are high and the total ion yield was also sufficient.

[0193] Similar behaviour was observed in a pointing mode of operation, as shown in the parameter optimization plots of Figs. 22A-D which show the total ion counts and the concordance correlation coefficients between the rapid evaporative ionization mass spectrometry imaging and iKnife technology reference spectra at different operating frequencies and voltages. A difference between a pointing and a cutting mode of operation is the time the electrosurgical tip of the sampling probe 21 is in contact with the same part of tissue. In a cutting mode of operation, the tip is constantly moving and therefore continuously touches fresh tissue, whereas the tip remains at the same tissue spot for a defined amount of time in a pointing mode of operation. This leads to longer exposure of heat, thus voltage and frequency have to be chosen in a way that carbonization is kept at a minimum. At the same time, longer exposure also creates more ions, decreasing the need for higher voltages to gain a sufficiently high TIC. By decreasing the time the tip remained about 1 mm inside the sample to a value of about 0.1 s, the exposure could be successfully decreased so that burn crater diameter was about 500 µm while providing good TICs and concordance correlation coefficients at about 2 kV and about 40 kHz.

[0194] The impact of heat exposure on the mass spectral pattern is shown in Figs. 23A-C. Figs. 23A-C illustrate changes in mass spectral patterns of porcine liver obtained in cutting mode for high (Fig. 23A) and low (Fig. 23B) voltages compared to an iKnife technology reference spectrum (Fig. 23C). There is a prominent peak in all mass spectra at m/z = 885.5 which is identified as a phosphatidyl-inositol species [PI(38:4)-H]-.

[0195] The iKnife technology reference mass spectrum shown in Fig. 23C shows the highest TIC together with the most distinct intensity difference between the PI peak and all other phospholipid signals. The signal to noise ratio decreases with increasing voltage, which particularly impacts the spectral pattern in the mass range between about m/z 600 and 1000, used for classification. Although the intensity difference between the PI peak and all other peaks is larger for the 2 kV (Fig. 23B) compared to the 6 kV spectrum (Fig. 23A), the TIC of the 2 kV spectrum is lower, indicating a lower level of chemical noise.

[0196] Optimized cutting and pointing mode parameters were used to analyse various types of tissues from different animals, including porcine and lamb liver, porcine kidney cortex and chicken skeletal muscle. Additionally, all samples were analysed by proper electrosurgical equipment ('iKnife' technology setup) to ensure selected experimental rapid evaporative ionization mass spectrometry imaging parameters are suitable for multiple tissue types. Principal component analysis of the data showed that the overall variance is mostly associated with the tissue types, not the modes of analysis (see Fig. 24). This demonstrates that the experimental parameters are universally applicable to various tissue types in terms of matching the iKnife technology reference mass spectral patterns.

Imaging liver with metastatic tumour



[0197] The imaging capability of the novel rapid evaporative ionization mass spectrometry platform (i.e. ion imager) was studied using human liver tumour samples (as illustrated in Fig. 25). For demonstration of the versatility of the platform a cutting mode rapid evaporative ionization mass spectrometry image was obtained on a first instrument whilst a pointing mode image was obtained on a Time of Flight mass spectrometer. Spatially resolved mass spectrometric information was co-registered with H&E images to locate mass spectra with the desired histological identity. Supervised multivariate analysis of the tissues revealed clear distinction between healthy and cancerous tissue for both rapid evaporative ionization mass spectrometry imaging and Desorption Electrospray Ionization ("DESI") imaging data.

[0198] The Desorption Electrospray Ionization ("DESI") images show a sharp border between the two tissue types as a result of the high spatial resolution and small pixel size of 100 µm. The upper half of the cutting mode rapid evaporative ionization mass spectrometry image contains pixels of mixed healthy and tumour pattern influences causing a blurred border. A possible explanation is due to the direction of the rapid evaporative ionization mass spectrometry cut that was performed which started at healthy tissue and continued towards the tumour region. This might have caused transport of tumour tissue pieces into the healthy area. Another reason may be inhomogeneous tissue below the surface of the seemingly cancerous area.

[0199] Assuming that the mass spectra are to be used as reference data for the iKnife technology, then only pixels with a high class-membership probability should be used for training the multivariate models (i.e. the sample classification model).

[0200] Unsupervised principal component analysis (PCA) demonstrates high intra-tissue-type spectral similarity together with spatially distinct clustering of healthy and cancerous data points in PCA space (see Fig. 26).

[0201] Desorption Electrospray Ionization ("DESI") imaging data acquired at high spatial resolution can also be used to locate histological fine structures and their corresponding mass spectra which can then be co-registered with the rapid evaporative ionization mass spectrometry data. A limiting factor for co-registration of Desorption Electrospray Ionization ("DESI") and rapid evaporative ionization mass spectrometry data is the spatial resolution currently achievable with the preferred rapid evaporative ionization mass spectrometry platform. While the cutting mode image was recorded at 500 µm pixel size, the pointing mode image features 750 µm sized pixels. In the case of this liver metastasis sample, the resolution is sufficient. However, in case of tissues with higher heterogeneity, higher spatial resolution images may be advantageous. The spatial resolution may be increased to decrease the diameter of the electrosurgical tip of the sampling probe 21 which would also be accompanied by lower spectral intensities. However, by connecting the sampling probe directly to the mass spectrometer inlet capillary (as is also done in the bipolar forceps approach described above) ion yield improves, thus overcoming the possible sensitivity issue. This also allows less penetration in z-direction, decreasing the probability of ionizing unanticipated tissue types.

[0202] Multivariate analysis of the liver metastasis samples shows a clear distinction of tissue types based on their molecular ion patterns. While rapid evaporative ionization mass spectrometry and Desorption Electrospray Ionization ("DESI") exhibit different ionization mechanisms resulting in mass spectrometric patterns that are not directly comparable to each other, univariate biochemical comparison of single ions provides a comparable measure for Desorption Electrospray Ionization ("DESI") and rapid evaporative ionization mass spectrometry co-registration. For certain compounds, the relative intensity difference between two tissue types is similar across all tissue types, ionization techniques and rapid evaporative ionization mass spectrometry analysis modes (cutting and pointing modes). This enables Desorption Electrospray Ionization ("DESI") to be used as a fold-change intensity-predictor for rapid evaporative ionization mass spectrometry based on up- and down-regulated compounds, which ultimately represents additional information for unknown tissue type identification. The higher spatial resolution of Desorption Electrospray Ionization ("DESI") allows the up- and down-regulated ions to be registered with certain histological features which may not be resolvable by rapid evaporative ionization mass spectrometry. This gives insight to the underlying histological composition of a tissue if certain changes in single ion intensities are observed in low resolution rapid evaporative ionization mass spectrometry.

[0203] In the case of metastatic liver comparison, two different phosphatidyl-ethanolamine (PE) species were found to possess opposite relative intensities between healthy and metastatic tissue types as shown in Fig. 27. The represented images are ion images of the two PE ion species. PE(38:4) has a higher abundance in healthy tissue in all four cases, with the rapid evaporative ionization mass spectrometry cutting mode image showing barely any presence of this ion in tumour tissue. However, compared to the Desorption Electrospray Ionization ("DESI") images where this lipid is well abundant even in tumour tissue, the absence of intensity has to be associated with the lower sensitivity achieved by rapid evaporative ionization mass spectrometry cutting. Opposite behaviour is seen by the ion [PE(36:1)-H]- showing elevated intensities in tumour tissue.

[0204] Future research will be dedicated to the comparison of multiple samples to obtain cross-validated relative intensity levels for ions of interest. Once enough data is collected, Desorption Electrospray Ionization ("DESI") can serve as a biochemical blueprint, allowing tissue types to be histologically annotated with higher confidence when analysed by rapid evaporative ionization mass spectrometry.

[0205] The ion imager may include a monopolar device with a separate return electrode or a bipolar device. Other arrangements are also contemplated in which the ion imager may include a multi-phase or 3-phase device and may include, for example, three or more separate electrodes or probes.

Setting up high throughput culturing, DNA isolation and MS data acquisition, determination of minimum culturing time



[0206] A customised Tecan EVO (RTM) platform incorporating automated colony imaging and colony picking was used to provide a reproducible system for high throughput workflows utilising rapid evaporative ionization mass spectrometry analysis. Using an automated platform helps minimise user time and errors to ensure the data is accurate and reproducible.

[0207] Automated rapid evaporative ionization mass spectrometry analysis was compared to the spectral profiles obtained using forceps. Five isolates of thirty species were examined using both methods and were also tested with and without the introduction of isopropanol ("IPA") matrix.

[0208] According to various embodiments a matrix (IPA) is added to the aerosol, smoke or vapour generated by the first device. The matrix is added to the aerosol, smoke or vapour prior to the aerosol, smoke or vapour impacting upon a collision surface.

[0209] It was apparent that for some bacterial species the Tecan (RTM) method generated noisy spectra. For example, Streptococcus pneumoniae generally produced noisy spectra with low intensities (see Fig. 28). Although some lipids could be observed, this was not reproducible.

[0210] Spectral profiles including both high and low mass lipids were observed for Fusobacterium nucleatum, but typically the profiles lacked those within higher mass ranges as in the mass spectrum shown in Fig. 29C. However, as shown by the spectra in Fig. 29A and Fig. 29B, these higher mass components were sometimes apparent and thus it is clear that, with optimisation, good quality spectra may be acquired.

[0211] Although a thorough analysis of each species needs to be performed, it was clear that the Tecan (RTM) produced data that encompasses higher mass range lipids. For example, as shown in Figs. 30A-D automated rapid evaporative ionization mass spectrometry produced a higher signal to noise ratio for Staphylococcus hominis.

[0212] The infusion of IPA, although producing peaks of significantly higher intensities, may result in the loss of higher mass range lipids as shown by the mass spectra in Fig. 30C and Fig. 30D in the case of S. hominis and Figs. 31A-B in the case of Pseudomonas aeruginosa. The presence of IPA also seems to increase the quality and reproducibility of the analysis. The statistical differentiation of strains appears to be equally efficient with and without IPA. Nevertheless, the high mass ranges seem to contribute to the separation of the strains in the dry mode, suggesting that, without being required for the separation, they might still bear valuable information.

[0213] It is also envisioned that a high-throughput sequencing pipeline may be implemented to attach the 'Gold' standard of taxonomic classification (16S rRNA gene sequence for bacteria and ITS region sequence for fungi) to each isolate rapid evaporative ionization mass spectrometry fingerprint. For instance, a filtration based platform such as the QIAGEN QIAcube that can process 96 isolates may be adapted to encompass the breath of clinical microbiology. Various different automated capillary electrophoresis technologies may be used to ensure PCR have successfully been generated. It is also contemplated that agarose gel electrophoresis may be used. A bioinformatic pipeline may be developed to allow for the automated analysis of sequence data and taxonomic classification against established sequence databases.

[0214] Many of the techniques described above are presented in the context of utilising rapid evaporative ionization mass spectrometry as an ionisation method. However, it will be appreciated that the techniques and apparatus described herein are not limited to rapid evaporative ionization mass spectrometry devices and may also be extended to other ambient ion sources and other methods of ambient ionisation. For example, a tool having fenestrations or aspiration ports may be provided as part of a laser surgery probe for aspirating aerosol, smoke or vapour generated using the laser. Further details of known ambient ion sources that may be suitable for use with the techniques and apparatus described herein are presented above.

Methods of medical treatment, surgery and diagnosis and non-medical methods



[0215] Various different embodiments are contemplated. According to some embodiments the methods disclosed above may be performed on in vivo, ex vivo or in vitro tissue. The tissue may comprise human or non-human animal tissue.

[0216] Various surgical, therapeutic, medical treatment and diagnostic methods are contemplated.

[0217] However, other embodiments are contemplated which relate to non-surgical and non-therapeutic methods of mass spectrometry and/or ion mobility spectrometry which are not performed on in vivo tissue. Other related embodiments are contemplated which are performed in an extracorporeal manner such that they are performed outside of the human or animal body.

[0218] Further embodiments are contemplated wherein the methods are performed on a non-living human or animal, for example, as part of an autopsy procedure.

[0219] Although the present invention has been described with reference to preferred embodiments, it will be understood by those skilled in the art that various changes in form and detail may be made without departing from the scope of the invention as set forth in the accompanying claims.


Claims

1. A method of ion imaging comprising:

automatically sampling a plurality of different locations on a sample using a first device arranged and adapted to generate aerosol, smoke or vapour from the sample, wherein said first device comprises a laser device;

automatically translating said sample relative to said first device before and/or during and/or after obtaining mass spectral data and/or ion mobility data from at least some of said locations on said sample;

providing a collision surface located within a vacuum chamber of a mass spectrometer and/or ion mobility spectrometer so as to generate analyte ions;

adding a matrix to said aerosol, smoke or vapour prior to said aerosol, smoke or vapour impacting upon said collision surface;

passing said aerosol, smoke or vapour into the vacuum chamber of the mass spectrometer and/or ion mobility spectrometer and causing at least some of said aerosol, smoke or vapour and matrix to impact upon said collision surface wherein at least some of said aerosol, smoke or vapour is ionized upon impacting said collision surface so as to generate analyte ions;

obtaining mass spectral data and/or ion mobility data corresponding to each said location; and

using said obtained mass spectral data and/or ion mobility data to construct, train or improve a sample classification model;

wherein said matrix is selected from the group consisting of: (i) a solvent for said aerosol, smoke or vapour; (ii) an organic solvent; (iii) a volatile compound; (iv) polar molecules; (v) water; (vi) one or more alcohols; (vii) methanol; (viii) ethanol; (ix) isopropanol; (x) acetone; (xi) acetonitrile; (xii) 1-butanol; (xiii) tetrahydrofuran; (xiv) ethyl acetate; (xv) ethylene glycol; (xvi) dimethyl sulfoxide; (xvii) an aldehyde; (xviii) a ketone; (xiv) non-polar molecules; (xx) hexane; (xxi) chloroform; (xxii) butanol; and (xxiii) propanol or wherein said matrix comprises a lockmass or calibration compound.


 
2. A method as claimed in claim 1, wherein said sample comprises a biological sample, biological tissue, human tissue, animal tissue, biological matter, a bacterial colony, a fungal colony or one or more bacterial strains.
 
3. A method as claimed in claim 1 or 2, wherein said sample comprises native or unmodified sample material, optionally wherein said native or unmodified sample material is unmodified by the addition of a matrix or reagent.
 
4. A method as claimed in any of claims 1, 2 or 3, wherein said sample classification model comprises a biological sample classification model, a biological tissue classification model, a human tissue classification model, an animal tissue classification model or a bacterial strain classification model.
 
5. A method as claimed in any preceding claim, further comprising constructing, training or improving said sample classification model in order either: (i) to distinguish between healthy and diseased tissue; (ii) to distinguish between potentially cancerous and non-cancerous tissue; (iii) to distinguish between different types or grades of cancerous tissue; (iv) to distinguish between different types or classes of sample material; (v) to determine whether or not one or more desired or undesired substances are present in said sample; (vi) to confirm the identity or authenticity of said sample; (vii) to determine whether or not one or more impurities, illegal substances or undesired substances are present in said sample; (viii) to determine whether a human or animal patient is at an increased risk of suffering an adverse outcome; (ix) to make or assist in the making a diagnosis or prognosis; and (x) to inform a surgeon, nurse, medic or robot of a medical, surgical or diagnostic outcome.
 
6. A method as claimed in any preceding claim, wherein the step of using said obtained mass spectral data and/or ion mobility data to construct, train or improve said sample classification model comprises performing a supervised or unsupervised multivariate statistical analysis of said mass spectral data and/or ion mobility data, optionally wherein said multivariate statistical analysis is selected from the group consisting of: (i) principal component analysis ("PCA"); and (ii) linear discriminant analysis ("LDA").
 
7. A method as claimed in any preceding claim, further comprising heating said collision surface optionally to a temperature selected from the group consisting of: (i) 200-300 °C; (ii) 300-400 °C; (iii) 400-500 °C; (iv) 500-600 °C; (v) 600-700 °C; (vi) 700-800 °C; (vii) 800-900 °C; (viii) 900-1000 °C; (ix) 1000-1100 °C; and (x) > 1100 °C.
 


Ansprüche

1. Verfahren zur lonenbildgebung, umfassend:

automatische Probenahme einer Vielzahl von verschiedenen Stellen auf einer Probe unter Verwendung einer ersten Vorrichtung, die so angeordnet und angepasst ist, dass sie Aerosol, Rauch oder Dampf aus der Probe erzeugt, wobei die erste Vorrichtung eine Laservorrichtung umfasst;

automatisches Verschieben der Probe relativ zu der ersten Vorrichtung vor und/oder während und/oder nach dem Erhalten von Massenspektraldaten und/oder lonenmobilitätsdaten von mindestens einigen der Stellen auf der Probe;

Bereitstellen einer Kollisionsfläche, die sich in einer Vakuumkammer eines Massenspektrometers und/oder lonenmobilitätsspektrometers befindet, um Analytionen zu erzeugen;

Hinzufügen einer Matrix zu dem Aerosol, dem Rauch oder dem Dampf, bevor das Aerosol, der Rauch oder der Dampf auf die Kollisionsoberfläche auftrifft;

Leiten des Aerosols, Rauchs oder Dampfes in die Vakuumkammer des Massenspektrometers und/oder lonenmobilitätsspektrometers und Bewirken, dass zumindest ein Teil des Aerosols, Rauchs oder Dampfes und der Matrix auf die Kollisionsoberfläche auftrifft, wobei zumindest ein Teil des Aerosols, Rauchs oder Dampfes beim Auftreffen auf die Kollisionsoberfläche ionisiert wird, um so Analytionen zu erzeugen;

Gewinnen von Massenspektraldaten und/oder lonenmobilitätsdaten, die jeder der Stelle entsprechen; und

Verwenden der gewonnenen Massenspektraldaten und/oder lonenmobilitätsdaten, um ein Probenklassifizierungsmodell zu konstruieren, zu schulen oder zu verbessern;

wobei die Matrix ausgewählt wird aus der Gruppe, die besteht aus: (i) einem Lösungsmittel für das Aerosol, den Rauch oder den Dampf; (ii) einem organischen Lösungsmittel; (iii) einer flüchtigen Verbindung; (iv) polaren Molekülen; (v) Wasser; (vi) einem oder mehreren Alkoholen; (vii) Methanol; (viii) Ethanol; (ix) Isopropanol; (x) Aceton; (xi) Acetonitril; (xii) 1-Butanol; (xiii) Tetrahydrofuran; (xiv) Ethylacetat; (xv) Ethylenglykol; (xvi) Dimethylsulfoxi d; (xvii) einem Aldehyd; (xviii) einem Keton; (xiv) unpolaren Molekülen; (xx) Hexan; (xxi) Chloroform; (xxii) Butanol; und (xxiii) Propanol oder wobei die Matrix eine Sperrmassen- oder Kalibrierungsverbindung umfasst.


 
2. Verfahren nach Anspruch 1, wobei die Probe eine biologische Probe, biologisches Gewebe, menschliches Gewebe, tierisches Gewebe, biologische Materie, eine Bakterienkolonie, eine Pilzkolonie oder einen oder mehrere Bakterienstämme umfasst.
 
3. Verfahren nach Anspruch 1 oder 2, wobei die Probe natives oder unmodifiziertes Probenmaterial umfasst, optional wobei das native oder unmodifizierte Probenmaterial nicht durch die Zugabe einer Matrix oder eines Reagens modifiziert ist.
 
4. Verfahren nach einem der Ansprüche 1, 2 oder 3, wobei das Probenklassifikationsmodell ein biologisches Probenklassifikationsmodell, ein biologisches Gewebeklassifikationsmodell, ein menschliches Gewebeklassifikationsmodell, ein tierisches Gewebeklassifikationsmodell oder ein Bakterienstammklassifikationsmodell umfasst.
 
5. Verfahren nach einem der vorstehenden Ansprüche, weiter umfassend ein Konstruieren, Schulen oder Verbessern des Probenklassifikationsmodells, um entweder: (i) zwischen gesundem und krankem Gewebe zu unterscheiden; (ii) zwischen potentiell krebsartigem und nicht-krebsartigem Gewebe zu unterscheiden; (iii) zwischen verschiedenen Arten oder Graden von krebsartigem Gewebe zu unterscheiden; (iv) zwischen verschiedenen Arten oder Klassen von Probenmaterial zu unterscheiden; (v) zu bestimmen, ob eine oder mehrere erwünschte oder unerwünschte Substanzen in der Probe vorhanden sind oder nicht; (vi) die Identität oder Authentizität der Probe zu bestätigen; (vii) zu bestimmen, ob eine oder mehrere Verunreinigungen, illegale Substanzen oder unerwünschte Substanzen in der Probe vorhanden sind oder nicht; (viii) zu bestimmen, ob ein menschlicher oder tierischer Patient einem erhöhten Risiko ausgesetzt ist, negative Auswirkungen zu erleiden; (ix) eine Diagnose oder Prognose zu stellen oder die Erstellung einer solchen zu unterstützen; und (x) einen Chirurgen, eine Krankenschwester, einen Sanitäter oder einen Roboter über ein medizinisches, chirurgisches oder diagnostisches Ergebnis zu informieren.
 
6. Verfahren nach Anspruch 1, wobei der Schritt des Verwendens der gewonnenen Massenspektraldaten und/oder Ionenmobilitätsdaten zum Konstruieren, Schulen oder Verbessern des Probenklassifizierungsmodells ein Durchführen einer überwachten oder unbeaufsichtigten multivariaten statistischen Analyse der Massenspektraldaten und/oder lonenmobilitätsdaten umfasst, wobei die multivariate statistische Analyse wahlweise aus der Gruppe ausgewählt wird, die besteht aus: (i) Hauptkomponentenanalyse ("PCA"); und (ii) linearer Diskriminanzanalyse ("LDA").
 
7. Verfahren nach einem der vorstehenden Ansprüche, weiter umfassend ein Erwärmen der Kollisionsoberfläche optional auf eine Temperatur, die ausgewählt wird aus der Gruppe bestehend aus: (i) 200-300 °C; (ii) 300-400 °C; (iii) 400-500 °C; (iv) 500-600 °C; (v) 600-700 °C; (vi) 700-800 °C; (vii) 800-900 °C; (viii) 900-1000 °C; (ix) 1000-1100 °C; und (x) > 1100 °C.
 


Revendications

1. Procédé d'imagerie ionique comprenant les étapes consistant à :

échantillonner automatiquement une pluralité d'emplacements différents sur un échantillon en utilisant un premier dispositif agencé et conçu pour générer un aérosol, une fumée ou une vapeur à partir de l'échantillon, dans lequel ledit premier dispositif comprend un dispositif laser ;

translater automatiquement ledit échantillon par rapport audit premier dispositif avant et/ou pendant et/ou après l'obtention de données spectrales de masse et/ou de données de mobilité ionique à partir d'au moins certains desdits emplacements sur ledit échantillon ;

fournir une surface de collision située dans une chambre à vide d'un spectromètre de masse et/ou d'un spectromètre de mobilité ionique de manière à générer des ions d'analyte ;

ajouter une matrice audit aérosol, à ladite fumée ou à ladite vapeur avant que ledit aérosol, ladite fumée ou ladite vapeur vienne frapper ladite surface de collision ;

faire passer ledit aérosol, ladite fumée ou ladite vapeur dans la chambre à vide du spectromètre de masse et/ou spectromètre de mobilité ionique et amener au moins une partie dudit aérosol, de ladite fumée ou de ladite vapeur et la matrice à venir frapper ladite surface de collision, dans lequel au moins une partie dudit aérosol, de ladite fumée ou de ladite vapeur est ionisée lors de l'impact sur ladite surface de collision de manière à générer des ions d'analyte ;

obtenir des données spectrales de masse et/ou données de mobilité ionique correspondant à chacun desdits emplacements ; et

utiliser lesdites données spectrales de masse et/ou données de mobilité ionique obtenues pour construire, former ou améliorer un modèle de classification d'échantillon ;

dans lequel ladite matrice est choisie parmi le groupe constitué par : (i) un solvant pour ledit aérosol, ladite fumée ou ladite vapeur; (ii) un solvant organique; (iii) un composé volatil ; (iv) des molécules polaires ; (v) de l'eau ; (vi) un ou plusieurs alcools ; (vii) méthanol ; (viii) éthanol ; (ix) isopropanol ; (x) acétone ; (xi) acétonitrile ; (xii) 1-butanol ; (xiii) tétrahydrofuranne ; (xiv) acétate d'éthyle ; (xv) éthylène glycol ; (xvi) diméthylsulfoxyde ; (xvii) un aldéhyde ; (xviii) une cétone ; (xiv) molécules non-polaires ; (xx) hexane ; (xxi) chloroforme ; (xxii) butanol ; et (xxiii) propanol ou dans lequel ladite matrice comprend une masse de référence ou un composé d'étalonnage.


 
2. Procédé selon la revendication 1, dans lequel ledit échantillon comprend un échantillon biologique, un tissu biologique, un tissu humain, un tissu animal, une matière biologique, une colonie de bactéries, une colonie de champignons ou une ou plusieurs souches bactériennes.
 
3. Procédé selon la revendication 1 ou 2, dans lequel ledit échantillon comprend un matériel d'échantillon natif ou non modifié, facultativement dans lequel ledit matériel d'échantillon natif ou non modifié est non modifié par l'ajout d'une matrice ou d'un réactif.
 
4. Procédé selon l'une quelconque des revendications 1, 2 ou 3, dans lequel ledit modèle de classification d'échantillon comprend un modèle de classification d'échantillon biologique, un modèle de classification de tissu biologique, un modèle de classification de tissu humain, un modèle de classification de tissu animal ou un modèle de classification de souche bactérienne.
 
5. Procédé selon l'une quelconque des revendications précédentes, comprenant en outre la construction, la formation ou l'amélioration dudit modèle de classification d'échantillon afin de soit : (i) faire la distinction entre un tissu sain et malade ; (ii) faire la distinction entre un tissu potentiellement cancéreux et non-cancéreux; (iii) faire la distinction entre différents types ou grades de tissu cancéreux; (iv) faire la distinction entre différents types ou classes de matériel d'échantillon ; (v) déterminer si une ou plusieurs substances souhaitées ou non souhaitées sont présentes ou non dans ledit échantillon ; (vi) confirmer l'identité ou l'authenticité dudit échantillon ; (vii) déterminer si une ou plusieurs impuretés, substances illégales ou substances non souhaitées sont présentes ou non dans ledit échantillon ; (viii) déterminer si un patient humain ou animal court un risque accru de souffrir d'un résultat néfaste ; (ix) établir ou aider à établir un diagnostic ou un pronostic; et (x) informer un chirurgien, une infirmière, un infirmier ou un robot d'un résultat médical, chirurgical ou diagnostique.
 
6. Procédé selon l'une quelconque des revendications précédentes, dans lequel l'étape d'utilisation desdites données spectrales de masse et/ou données de mobilité ionique obtenues pour construire, former ou améliorer ledit modèle de classification d'échantillon comprend la réalisation d'une analyse statistique multivariée supervisée ou non supervisée desdites données spectrales de masse et/ou données de mobilité ionique, facultativement dans lequel ladite analyse statistique multivariée est choisie parmi le groupe constitué de : (i) une analyse en composantes principales (« PCA ») ; et (ii) une analyse discriminante linéaire (« LDA »).
 
7. Procédé selon l'une quelconque des revendications précédentes, comprenant en outre le chauffage de ladite surface de collision facultativement jusqu'à une température choisie parmi le groupe constitué de : (i) 200 à 300°C ; (ii) 300 à 400°C ; (iii) 400 à 500°C ; (iv) 500 à 600°C ; (v) 600 à 700°C ; (vi) 700 à 800°C ; (vii) 800 à 900°C ; (viii) 900 à 1 000°C ; (ix) 1 000 à 1 100°C ; et (x) > 1 100°C.
 




Drawing























































































REFERENCES CITED IN THE DESCRIPTION



This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

Non-patent literature cited in the description