System Maintenance Notice

Please note that maintenance work is scheduled between 4 AM and 5 AM - CET. During this time, you may experience reduced performance. Thanks for your patience.

Publications

English Abstract Journal Article Ahead of print

PubMed: 41577495 DOI: 10.1016/j.annpat.2025.12.009

[Open data: main objective achieved for the French Society of Pathology's data challenges].

Ann Pathol

Authors: Philippe Bertheau, Laurie Alla, Agathe Delaune, Nicolas Loménie, Amir Khammari, Eve Maubec, Frédéric Staroz, Sophie Prévot

Affiliations

  • Service de pathologie, hôpital Saint-Louis, GHU Nord AP-HP et université Paris Cité, Paris, France.
  • Health Data Hub, 9, rue Georges-Pitard, 75015 Paris, France.
  • Université Paris Cité, LIPADE, 75006 Paris, France.
  • Département de dermatologie, CIC 1413, INSERM UMR 1302/EMR6001 INCIT, CHU de Nantes, Nantes université, Nantes, France.
  • Assistance publique Hôpitaux de Paris, hôpital Avicenne, université Sorbonne Paris Nord, Bobigny, université Paris Cité, inserm UMR 1124, Paris, France.
  • Ouest pathologie, Quimper, France.
  • Service de Pathologie hôpital de Bicêtre, GHU Paris Saclay et université Paris Saclay, Le Kremlin Bicêtre, France.

Abstract

The French Society of Pathology organized two data challenges, with the support of the Health Data Hub: TissueNet in 2020, focusing on cervical lesions, and Visiomel in 2023 dedicated to melanoma in partnership with the French Society of Dermatology and the National Professional Council of Pathologists. The datasets were open a few weeks to competitors worldwide during each challenge. However, these data were then closed, awaiting a complete and permanent opening, known as open data. This ultimate step allows free access to the data to promote global research programs in these diseases. The data from the two data challenges have been available as open data on data.gouv.fr since January 2025 for TissueNet (https://www.data.gouv.fr/datasets/tissuenet-detect-lesions-in-uterine-cervix-specimens-open-data-set/) and since September 2025 for Visiomel (https://www.data.gouv.fr/datasets/visiomel-database-predicting-cutaneous-melanoma-relapse/). Making health data open is a demanding multi-step process - organizationally, technically, and regulatorily - and it is this process that we discuss here.

MeSH

Melanoma, Pathology, Wounds And Injuries, Cervix Uteri