The online medical appointment giant Doctolib has created its "clinical artificial intelligence (AI) laboratory", in connection with the University Hospital of Nantes, Inserm and various learned societies, and is investing 20 million euros in it by 2026, it announced Monday evening.
This "collective project", officially launched on Monday after "months" of work, brings together around Doctolib "reference" institutions in the field of AI including the National Institute for Research in Digital Science and Technology (INRIA), and the German DFKI, Doctolib told AFP, confirming information from the daily newspaper Le Figaro.
It also brings together teams from the University Hospital of Nantes, various healthcare professionals "using" the platform, several learned societies including the French Society of Pediatrics, and "leading laboratories in children's health such as Professor Mazza's laboratory in Lyon," the company specifies.
The research units are thus distributed between Paris, Nantes and Berlin.
The goal is to "build on this expertise" to eventually offer doctors clinical decision-making tools and patients "a complete health assistant" to prepare for their consultation, better understand their diagnosis or treatment, with "the greatest medical reliability" and "the safest safety standards," Doctolib emphasizes.
The tech giant points out that it is in contact with more than 400,000 healthcare professionals, 1,000 healthcare facilities, and 90 million patients. It has also recruited to bring the number of people working in research and development to 900, including 100 dedicated solely to AI.
Doctolib wants to develop tools that, unlike general-purpose AI assistants, only respond "when the level of trust is sufficient" and will be "trained on validated and local knowledge, not on the entire web," explained Doctolib's president, Stanislas Niox-Chateau, to Le Figaro.
According to Le Figaro, the Nantes University Hospital will, for example, contribute to establishing rules on the level of urgency in the care of children.
For services to doctors, "in the future, (...) we will go further in assisting with anamnesis (reconstructing the patient's medical history, editor's note), prescription or diagnosis. AI will also be able to help predict the risk of developing a disease," Mr. Niox-Chateau told Le Figaro.
The research will be published and made accessible to "promote innovation in health in Europe," the company stated.
English 
French