Mihaela Pop · Maxime Sermesant Pierre-Marc Jodoin · Alain Lalande Xiahai Zhuang · Guang Yang Alistair Young · Olivier Bernard (Eds.) Statistical Atlases 3 and Computational Models 6 6 0 of the Heart 1 S C N ACDC and MMWHS Challenges L 8th International Workshop, STACOM 2017 Held in Conjunction with MICCAI 2017 Quebec City, Canada, September 10–14, 2017, Revised Selected Papers 123 Lecture Notes in Computer Science 10663 Commenced Publication in 1973 Founding and Former Series Editors: Gerhard Goos, Juris Hartmanis, and Jan van Leeuwen Editorial Board David Hutchison Lancaster University, Lancaster, UK Takeo Kanade Carnegie Mellon University, Pittsburgh, PA, USA Josef Kittler University of Surrey, Guildford, UK Jon M. Kleinberg Cornell University, Ithaca, NY, USA Friedemann Mattern ETH Zurich, Zurich, Switzerland John C. Mitchell Stanford University, Stanford, CA, USA Moni Naor Weizmann Institute of Science, Rehovot, Israel C. Pandu Rangan Indian Institute of Technology Madras, Chennai, India Bernhard Steffen TU Dortmund University, Dortmund, Germany Demetri Terzopoulos University of California, Los Angeles, CA, USA Doug Tygar University of California, Berkeley, CA, USA Gerhard Weikum Max Planck Institute for Informatics, Saarbrücken, Germany More information about this series at http://www.springer.com/series/7412 Mihaela Pop Maxime Sermesant (cid:129) Pierre-Marc Jodoin Alain Lalande (cid:129) Xiahai Zhuang Guang Yang (cid:129) Alistair Young Olivier Bernard (Eds.) (cid:129) Statistical Atlases and Computational Models of the Heart ACDC and MMWHS Challenges 8th International Workshop, STACOM 2017 Held in Conjunction with MICCAI 2017 – Quebec City, Canada, September 10 14, 2017 Revised Selected Papers 123 Editors Mihaela Pop XiahaiZhuang Sunnybrook Research Institute FudanUniversity University of Toronto Shanghai Toronto, ON China Canada Guang Yang Maxime Sermesant Cardiovascular Research Center, National Inria-Asclepios Heart andLungInstitute SophiaAntipolis RoyalBrompton Hospital France London UK Pierre-Marc Jodoin UniversitédeSherbrooke Alistair Young Quebec City,QC University of Auckland Canada Auckland NewZealand AlainLalande University Bourgogne Olivier Bernard Dijon CREATIS, INSA-Lyon France Villeurbanne France ISSN 0302-9743 ISSN 1611-3349 (electronic) Lecture Notesin Computer Science ISBN 978-3-319-75540-3 ISBN978-3-319-75541-0 (eBook) https://doi.org/10.1007/978-3-319-75541-0 LibraryofCongressControlNumber:2018935903 LNCSSublibrary:SL6–ImageProcessing,ComputerVision,PatternRecognition,andGraphics ©SpringerInternationalPublishingAG,partofSpringerNature2018 Thisworkissubjecttocopyright.AllrightsarereservedbythePublisher,whetherthewholeorpartofthe material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storageandretrieval,electronicadaptation,computersoftware,orbysimilarordissimilarmethodologynow knownorhereafterdeveloped. Theuseofgeneraldescriptivenames,registerednames,trademarks,servicemarks,etc.inthispublication doesnotimply,evenintheabsenceofaspecificstatement,thatsuchnamesareexemptfromtherelevant protectivelawsandregulationsandthereforefreeforgeneraluse. Thepublisher,theauthorsandtheeditorsaresafetoassumethattheadviceandinformationinthisbookare believedtobetrueandaccurateatthedateofpublication.Neitherthepublishernortheauthorsortheeditors give a warranty, express or implied, with respect to the material contained herein or for any errors or omissionsthatmayhavebeenmade.Thepublisherremainsneutralwithregardtojurisdictionalclaimsin publishedmapsandinstitutionalaffiliations. Printedonacid-freepaper ThisSpringerimprintispublishedbytheregisteredcompanySpringerInternationalPublishingAG partofSpringerNature Theregisteredcompanyaddressis:Gewerbestrasse11,6330Cham,Switzerland Preface Integrative models of cardiac function are important for understanding disease, eval- uating treatment, and planning intervention. In recent years, there has been consider- able progress in cardiac image analysis techniques, cardiac atlases, and computational models, which can integrate data from large-scale databases of heart shape, function, and physiology. However, significant clinical translation of these tools is constrained by the lack of complete and rigorous technical and clinical validation, as well as benchmarkingofthedevelopedtools.Tothisend,commonandavailableground-truth data capturing generic knowledge on the healthy and pathological heart are required. SeveraleffortsarenowestablishedtoprovideWeb-accessiblestructuralandfunctional atlases of the normal and pathological heart for clinical, research, and educational purposes.Webelievethattheseapproacheswillonlybeeffectivelydevelopedthrough collaboration across the full research scope of the cardiac imaging and modeling communities. STACOM 2017 was held in conjunction with the MICCAI 2017 international conference(QuebecCity, Canada), following thepast seven editions:STACOM 2010 (2010, Beijing, China), STACOM 2011 (Toronto, Canada), STACOM 2012 (Nice, France),STACOM2013(Nagoya,Japan),STACOM2014(Boston,USA),STACOM 2015 (Munich, Germany), and STACOM 2016 (Athens, Greece). STACOM 2017 provided a forum in which to discuss the latest developments in various areas of computational imaging and modeling of the heart as well as statistical cardiac atlases. The topics of the workshop included: cardiac imaging and image processing, atlas construction, statistical modeling of cardiac function across different patient popula- tions, cardiac computational physiology, model customization, atlas-based functional analysis,ontologicalschematafordataandresults,integratedfunctional andstructural analyses,aswellasthepre-clinicalandclinicalapplicabilityofthesemethods.Besides regular contributing papers, additional efforts of this year’s STACOM workshop were also focused on two challenges: ACDC and MM-WHS, described here. A total of 27 papers (regular papers and from the two challenges) were accepted to be presented at STACOM 2017, and are published in this LNCS proceedings volume. ACDC Automatic Cardiac Diagnostic Challenge The overarching objective of this challenge is two-fold: (1)TocomparetheperformanceofautomaticMRIsegmentationmethodsontheleft ventricular endocardium and epicardium as well as the right ventricular endocardium for both the end-diastolic and end-systolic phase instances (2) To compare the performance of automatic methods for the classification of examinations in five classes (normal case, myocardial infarction with altered left ventricular ejection fraction, dilated cardiomyopathy, hypertrophic cardiomyopathy, abnormal right ventricle) VI Preface TheoverallAC-DCdatasetcontainsrealclinicalexaminationsfrom150patientsall acquiredattheUniversityHospitalofDijon(France).Eachpatientdatasetcomeswith two ground truth information: (1) the pathology the patient suffers from and (2) a pixel-accurate delineation of each cardiac region (the endocardial wall of the left ventricle and of the right ventricles, and the epicardial wall of the left ventricle). The segmentation ground truthswere manually drawnbytwo experts (acardiologistanda nuclear medicine physician with experience in cardiology and MRI). The delineation was done for the most importantphases of the cardiac cycle, i.e., diastole and systole. The diastolic phase is the first image acquired after the R-wave of the ECG while the systolicphaseisthemomentwhenthemitralvalvereachesitsmaximumexcursionand the myocardium reaches its maximum contraction. The dataset as well as the results obtained by the challengers can be found here: http://acdc.creatis.insa-lyon.fr/. MM-WHS — Multi-Modal Whole-Heart Segmentation Challenge Accuratecomputing,modeling,andanalysisofthewhole-heartsubstructuresfrom3D medical image scans is important in the development of clinical applications. Seg- mentation and registration of whole-heart images is, however, still challenging. The extraction and modeling of whole-heart substructures currently relies heavily on manual delineation, which is a time-consuming task and is also prone to errors and dependent on the expertise of the observer; therefore, fully automated methods are highly desirable. Over the past decade, many techniques have been proposed to solve this ill-posed problem, particularly for whole-heart segmentation, such as atlas-based methods, statistical shape model-based methods, and recently emerged deep- learning-based methods. The organized MM-WHS Challenge provided an open and fair competition for various research groups to test and validate their methods, par- ticularly whole-heart segmentation, on datasets that were acquired in real clinical environments.TheaimoftheMM-WHSChallengewasnotonlytobenchmarkvarious whole-heart segmentation algorithms, but also to cover the topic of general cardiac image segmentation, registration, and modeling, and to raise discussions for further technical development and clinical deployment. The organizers provided more than 120 datasets from multiple sites, including 60 cardiac CT/CTA and 60 cardiac MRI volumes in 3D that cover whole-heart sub- structuresformulti-modalitywhole-heartsegmentation.Alltheseclinicaldatareceived institutional ethic approval and were anonymized. Both cardiac CT and cardiac MRI data were acquired in real clinical environment for patients that cover a wide range of cardiacdiseasesaswellasnormalcontrols.Wereceivedgreatinterestfromparticipants allovertheworldandtheproposedmethodshaveachievedsubstantialmethodological innovations and significant performance improvement. We aim at keeping the MM-WHS Challenge as a long-term event for participants who may not be able to enter the competition, but are interested in further developments. All the relevant informationandchallengeresultscanbefoundat:http://www.sdspeople.fudan.edu.cn/ zhuangxiahai/0/mmwhs/. Preface VII We would like to thank all organizers, reviewers, authors, and sponsors for their time, efforts, contributions, and financial support in making STACOM 2017 a suc- cessfulevent.Wehopethattheresultsobtainedbythesetwochallenges,alongwiththe regular paper contributions, will act to accelerate progress in the important areas of cardiac image analysis, heart function, and structure analysis. September 2017 Mihaela Pop Maxime Sermesant Pierre-Marc Jodoin Alain Lalande Xiahai Zhuang Guang Yang Alistair Young Olivier Bernard Organization Chairs and Organizers STACOM Mihaela Pop University of Toronto, Canada Alistair Young University of Auckland, New Zealand Maxime Sermesant Inria-Asclepios, France Tommaso Mansi Siemens, USA Kawal Rhode KCL, London, UK Kristin McLeod Simula, Norway ACDC Challenge Pierre-Marc Jodoin University of Sherbrook, Canada Alain Lalande University of Bourgogne, France Olivier Bernard University of Lyon, France MM-WHS Challenge Xiahai Zhuang Fudan University Guang Yang Imperial College London, UK Lei Li Shanghai Jiao Tong University, China OCS - Springer Conference Submission/Publication System Mihaela Pop Medical Biophysics, University of Toronto, Sunnybrook Research Institute, Toronto, Canada Webmaster Avan Suinesiaputra University of Auckland, New Zealand Workshop Website stacom2017.cardiacatlas.org X Organization Sponsors We are extremely grateful for the industrial funding support. The STACOM 2017 workshop received financial support from the following sponsors: SciMedia Ltd (http://www.scimedia.com/) for STACOM Imeka (http://imeka.ca) for ACDC challenge Nvidia (http://nvidia.com) for MM-WHS challenge Arterys (http://arterys.com) for MM-WHS challenge