Can AI predict epithelial lesion categories via automated analysis of cervical biopsies: The TissueNet challenge?

Autor:	Loménie N; LIPADE, UFR Mathématiques-Informatiques, Université Paris Cité, 45 rue des Saints-Pères, 75006 Paris, France., Bertrand C; Tribun Health, Courbevoie, France., Fick RHJ; Tribun Health, Courbevoie, France., Ben Hadj S; Tribun Health, Courbevoie, France., Tayart B; Idemia, Courbevoie, France., Tilmant C; GHICL, Lille, France., Farré I; XPath Nord, Leulinghem, France., Azdad SZ; Algoscope, 9 rue Gaspard Monge, 60200 Compiègne, France., Dahmani S; Algoscope, 9 rue Gaspard Monge, 60200 Compiègne, France., Dequen G; Laboratoire Modélisation, Information, Systèmes (MIS), Université de Picardie Jules Verne, 80080 Amiens, France., Feng M; Tongji University, Shanghai, China., Xu K; Tongji University, Shanghai, China., Li Z; Tongji University, Shanghai, China., Prevot S; Pathologie, CHU Bicêtre, APHP, 78 Rue du Général Leclerc, 94270 Le Kremlin-Bicêtre, France., Bergeron C; CerbaPath, 16 bis rue Odessa, Paris 75014, France., Bataillon G; Pathologie, Institut Curie, 26 rue d'Ulm, 75005 Paris, France., Devouassoux-Shisheboran M; Centre de Pathologie Sud des Hospices Civils de Lyon, Centre Hospitalier Lyon Sud, 165 chemin du grand Revoyet, 69495 Pierre Bénite Cedex, France., Glaser C; Pathologie, CHG Versailles, 177 Rue de Versailles, 78150 Le Chesnay-Rocquencourt, France., Delaune A; Plateforme de données de santé - Health Data Hub, 9 rue Georges Pitard, 75015 Paris, France., Valmary-Degano S; Pathologie, Université Grenoble Alpes, Inserm U1209, CNRS UMR5309, Institute for Advanced Biosciences, CHU, Grenoble 38000, France., Bertheau P; Pathologie, CHU Saint-Louis, APHP, Université Paris Cité, 1 avenue Claude Vellefaux, 75010 Paris, France.
Jazyk:	angličtina
Zdroj:	Journal of pathology informatics [J Pathol Inform] 2022 Oct 05; Vol. 13, pp. 100149. Date of Electronic Publication: 2022 Oct 05 (Print Publication: 2022).
DOI:	10.1016/j.jpi.2022.100149
Abstrakt:	The French Society of Pathology (SFP) organized its first data challenge in 2020 with the help of the Health Data Hub (HDH). The organization of this event first consisted of recruiting nearly 5000 cervical biopsy slides obtained from 20 pathology centers. After ensuring that patients did not refuse to include their slides in the project, the slides were anonymized, digitized, and annotated by expert pathologists, and finally uploaded to a data challenge platform for competitors from around the world. Competing teams had to develop algorithms that could distinguish 4 diagnostic classes in cervical epithelial lesions. Among the many submissions from competitors, the best algorithms achieved an overall score close to 95%. The final part of the competition lasted only 6 weeks, and the goal of SFP and HDH is now to allow for the collection to be published in open access for the scientific community. In this report, we have performed a "post-competition analysis" of the results. We first described the algorithmic pipelines of 3 top competitors. We then analyzed several difficult cases that even the top competitors could not predict correctly. A medical committee of several expert pathologists looked for possible explanations for these erroneous results by reviewing the images, and we present their findings here targeted for a large audience of pathologists and data scientists in the field of digital pathology. (© 2022 The Author(s).)
Databáze:	MEDLINE
Externí odkaz:	Zobrazit plný text záznamu