EP 3791310 A4 20220330 - SYSTEMS AND METHODS FOR PRE-PROCESSING ANATOMICAL IMAGES FOR FEEDING INTO A CLASSIFICATION NEURAL NETWORK

Title (en)

SYSTEMS AND METHODS FOR PRE-PROCESSING ANATOMICAL IMAGES FOR FEEDING INTO A CLASSIFICATION NEURAL NETWORK

Title (de)

SYSTEME UND VERFAHREN ZUR VORVERARBEITUNG ANATOMISCHER BILDER ZUR EINFÜHRUNG IN EIN NEURONALES KLASSIFIZIERUNGSNETZ

Title (fr)

SYSTÈMES ET PROCÉDÉS DE PRÉTRAITEMENT D'IMAGES ANATOMIQUES POUR INTRODUCTION DANS UN RÉSEAU NEURONAL DE CLASSIFICATION

Publication

EP 3791310 A4 20220330 (EN)

Application

EP 19800865 A 20190507

Priority

US 201815972912 A 20180507
US 201916269619 A 20190207
US 201916269633 A 20190207
IB 2019053726 W 20190507

Abstract (en)

[origin: EP3567525A1] There is provided a method comprising: providing two anatomical images 104 of a target individual, each captured at a unique orientation of the target individual, inputting first and second anatomical images respectively into a first and second convolutional neural network (CNN) of a classifier to respectively output first and second feature vectors, inputting a concatenation of the first and second feature vectors into a fully connected layer of the classifier 110, and computing an indication of distinct visual finding(s) 112 present in the anatomical images by the fully connected layer, wherein the statistical classifier is trained on a training dataset including two anatomical images of each respective sample individual, each image captured at a respective unique orientation of the target individual, and a tag created based on an analysis that maps respective individual sentences of a text based radiology report to one of multiple indications of visual findings.

IPC 8 full level

G06K 9/62 (2022.01)

CPC (source: EP US)

G06F 18/256 (2023.01 - EP US); G06V 10/454 (2022.01 - EP US); G06V 2201/03 (2022.01 - EP)

Citation (search report)

[Y] WO 2006055160 A2 20060526 - EASTMAN KODAK CO [US], et al
[A] US 2006110021 A1 20060525 - LUO HUI [US], et al
[Y] CHERMAN EVERTON ALVARES ET AL: "Multi-label Problem Transformation Methods: A Case Study", CLEI ELECTRONIC JOURNAL, 1 April 2011 (2011-04-01), pages 4 - 4, XP055890951, Retrieved from the Internet <URL:http://www.scielo.edu.uy/pdf/cleiej/v14n1/v14n1a05.pdf> [retrieved on 20220214], DOI: 10.19153/cleiej.14.1.4
[A] "BILDVERARBEITUNG IN DER MEDIZIN: ALGORITHMEN, SYSTEME, ANWENDUNGEN. PROCEEDINGS DES WORKSHOPS", 1 January 2005, ISBN: 978-3-540-25052-4, article MÜLLER HENNING ET AL: "Logo and Text Removal for Medical Image Retrieval", pages: 35 - 39, XP055890944, DOI: 10.1007/3-540-26431-0_8
[A] ZHU YINGXUAN ET AL: "An automatic system to detect and extract texts in medical images for de-identification", PROCEEDINGS OF SPIE, vol. 7628, 4 March 2010 (2010-03-04), 1000 20th St. Bellingham WA 98225-6705 USA, XP055890947, ISSN: 0277-786X, ISBN: 978-1-5106-4548-6, DOI: 10.1117/12.855588
[A] YUNCHAO WEI ET AL: "CNN: Single-label to Multi-label", 9 June 2014 (2014-06-09), pages 1 - 14, XP055404290, Retrieved from the Internet <URL:https://arxiv.org/pdf/1406.5726.pdf> [retrieved on 20170906], DOI: 10.1109/TPAMI.2015.2491929
See references of WO 2019215606A1

Designated contracting state (EPC)

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

DOCDB simple family (publication)

EP 3567525 A1 20191113; DE 202019005911 U1 20230419; EP 3791310 A1 20210317; EP 3791310 A4 20220330; EP 3791325 A1 20210317; EP 3791325 A4 20220413; JP 2019195627 A 20191114; WO 2019215604 A1 20191114; WO 2019215605 A1 20191114; WO 2019215606 A1 20191114

DOCDB simple family (application)

EP 19173136 A 20190507; DE 202019005911 U 20190507; EP 19800738 A 20190507; EP 19800865 A 20190507; IB 2019053724 W 20190507; IB 2019053725 W 20190507; IB 2019053726 W 20190507; JP 2019087284 A 20190507