The FFCD and Owkin Join Forces to Apply Machine Learning to Digestive Oncology


Tags: Partnership / ML


Date:March 5th, 2020


The FFCD and Owkin Join Forces to Apply Machine Learning to Digestive Oncology

Paris, March 5, 2020

Owkin – a startup specializing in machine learning applied to clinical research, and the FFCD – Fédération Francophone de Cancérologie Digestive, an academic cooperative group specializing in digestive cancers – announce the framework agreement of their collaboration taking place over the next 5 years on prospective cohorts of digestive cancer patients.

Machine learning for digestive oncology: understanding and predicting relapses

In close collaboration, Owkin’s data scientists and FFCD experts will develop machine learning models from FFCD databases of homogeneous, longitudinal, and multimodal cohorts from clinical trial protocols. To understand and predict relapse, Owkin’s teams will analyze data from 3,000 patients with stage II or III colorectal cancer. In parallel, Owkin will make its Owkin Studio research platform available to FFCD medical researchers. This will enable them to benefit from the best AI algorithms, while doctors will be able to collaborate directly with the startup’s data scientists and experts via Owkin Studio to interpret the results of the algorithms trained on their data. The pseudo-anonymized data on the FFCD servers, hosted at the University of Burgundy, will ensure patient confidentiality while the federated learning technologies developed by Owkin will permit the integration of the data into collaborative research projects. This technology further ensures the security and traceability of operations. From digestive oncology to multiple collaborations.

Together, Owkin and FFCD share the idea of future research projects on digestive cancers, in particular on the prediction of mutations from anatomo-pathology slides. The launch of these new projects is facilitated by the signing of the framework collaboration agreement. “Democratizing access to AI and machine learning for world-renowned experts and physicians demonstrate how this technology can contribute to the treatment of serious diseases. We are delighted to be able to collaborate in a collective effort with the best researchers in their field and ultimately improve patient care,” said Meriem Sefta, Head of Partnerships at Owkin. “AI in clinical research is a natural evolution of our structure. Clinical research in the 21st century will combine the historical skills of clinical physicians with the revolutionary technological approach of a young and enthusiastic company. The aim of this collaboration is to accelerate the optimization of therapeutic strategies in digestive oncology. “said Professor Pierre Michel, President of the FFCD.

About the Fédération Francophone de Cancérologie Digestive

FFCD is an academic cooperative group (Association dite “Loi de 1901”), specialized in digestive oncology, which has established, in complete independence, research and scientific communication projects or training in the fields of digestive cancers (physiopathology, diagnosis and therapy), since June 12, 1981. The French-speaking Federation of Digestive Oncology brings together all the players in the field of digestive oncology (gastroenterologists, surgeons, oncologists, radiotherapists, radiologists, pathologists) in more than 300 investigative centres throughout France. It has developed exceptional scientific expertise in cancer treatment by initiating and coordinating clinical and biological trials at the national and international levels. For more information, visit and follow @ffcd_cancerdig on Twitter.

About Owkin

Owkin specializes in artificial intelligence technologies applied to clinical research. It was co-founded in 2016 by Thomas Clozel, a haematologist oncologist and researcher, and Gilles Wainrib, a computer science teacher-researcher at the École Normale Supérieure, and a Stanford University PostDoc. Owkin enables researchers to use data from health care or research activities to train interpretable machine learning models. These models allow better prediction of patient prognosis and response to treatment, developing new generations of biomarkers. In October 2019, Owkin published its breakthrough analysis of tumour biology using an interpretable deep-learning model, called MesoNet, in Nature Medicine. Owkin is among the first companies to use Federated Learning technologies, a decentralised analysis approach that protects patient data by ensuring that it never leaves the hospital.

For more information, visit and follow @OWKINscience on Twitter.

Press Contacts at Owkin – Agence Raoul
Talia Lliteras : 07 87 21 81 90 –
Pierre Gatey : 06 07 33 89 45 –
Sibylle de Villeneuve : 06 45 29 58 57 –
Press Contacts at FFCD – Agence Concept Editions
Nathalie Deveza : 0625143133 –
Fabienne Roulleaux : 0602683609 –