Skip to Main content Skip to Navigation
Journal articles

Five simultaneous artificial intelligence data challenges on ultrasound, CT, and MRI

N. Lassau 1, 2, 3 T. Estienne 4 P. de Vomecourt 2 M. Azoulay 5 John Cagnol 6 G. Garcia 1 M. Majer 1 E. Jehanno 2 R. Renard-Penna 7 C. Balleyguier 1, 2 F. Bidault 8, 1, 3, 2 C. Caramella 1, 2 T. Jacques 9 F. Dubrulle 10 J. Behr 11 N. Poussange 12 J. Bocquet 13 S. Montagne 14 F. Cornelis 15 M. Faruch 16 S. Brunelle 17 B. Bresson 17 A. Jalaguier-Coudray 18 N. Amoretti 19 A. Blum 20 A. Paisant 21 V. Herreros O. Rouviere 22 S. Si-Mohamed 23 L. Di Marco 24 O. Hauger 25 M. Garetier 26 F. Pigneur 27 A. Bergère 2 C. Cyteval 28 L. Fournier 29, 30 C. Malhaire 31 J.-L. Drape 32 E. Poncelet 33 C. Bordonne 34 H. Cauliez 35 J.-F. Budzik 36 M. Boisserie 37 T. Willaume 38 S. Molière 38 N. Peyron Faure 39 S. Caius Giurca 40 V. Juhan 41 T. Caramella 42 A. Perrey 43 F. Desmots 44 M. Faivre-Pierre 45 M. Abitbol 46 R. Lotte 47 D. Istrati 48 D. Guenoun 49, 50 A. Luciani 51, 27 M. Zins 52 J.-F. Meder 53 A. Cotten 54
2 IR4M/U8081 éq.3 - Imagerie multimodale en cancérologie.
IR4M - Imagerie par Résonance Magnétique Médicale et Multi-Modalités, IGR - Institut Gustave Roussy
Abstract : PurposeThe goal of this data challenge was to create a structured dynamic with the following objectives: (1) teach radiologists the new rules of General Data Protection Regulation (GDPR), while building a large multicentric prospective database of ultrasound, computed tomography (CT) and MRI patient images; (2) build a network including radiologists, researchers, start-ups, large companies, and students from engineering schools, and; (3) provide all French stakeholders working together during 5 data challenges with a secured framework, offering a realistic picture of the benefits and concerns in October 2018.Materials and methodsRelevant clinical questions were chosen by the Société Francaise de Radiologie. The challenge was designed to respect all French ethical and data protection constraints. Multidisciplinary teams with at least one radiologist, one engineering student, and a company and/or research lab were gathered using different networks, and clinical databases were created accordingly.ResultsFive challenges were launched: detection of meniscal tears on MRI, segmentation of renal cortex on CT, detection and characterization of liver lesions on ultrasound, detection of breast lesions on MRI, and characterization of thyroid cartilage lesions on CT. A total of 5,170 images within 4 months were provided for the challenge by 46 radiology services. Twenty-six multidisciplinary teams with 181 contestants worked for one month on the challenges. Three challenges, meniscal tears, renal cortex, and liver lesions, resulted in an accuracy > 90%. The fourth challenge (breast) reached 82% and the lastone (thyroid) 70%.ConclusionTheses five challenges were able to gather a large community of radiologists, engineers, researchers, and companies in a very short period of time. The accurate results of three of the five modalities suggest that artificial intelligence is a promising tool in these radiology modalities.
Document type :
Journal articles
Complete list of metadata
Contributor : Accord Elsevier Ccsd Connect in order to contact the contributor
Submitted on : Friday, October 22, 2021 - 5:29:57 PM
Last modification on : Tuesday, January 4, 2022 - 6:17:49 AM


Files produced by the author(s)


Distributed under a Creative Commons Attribution - NonCommercial 4.0 International License



N. Lassau, T. Estienne, P. de Vomecourt, M. Azoulay, John Cagnol, et al.. Five simultaneous artificial intelligence data challenges on ultrasound, CT, and MRI. Diagnostic and Interventional Imaging, Elsevier, 2019, 100 (4), pp.199-209. ⟨10.1016/j.diii.2019.02.001⟩. ⟨hal-02528753⟩



Les métriques sont temporairement indisponibles