About

Computer vision @Owkin

I am currently working as a data scientist at Owkin. Owkin is an emerging biotech company with the mission to find the right treatment for every patient. At Owkin, we are leveraging cutting-edge machine learning to deliver better drugs and diagnostics. I am part of the Medical Imaging group which focuses on the analysis and representation of histopathology data. More specically, I am involved in the development and continuous improvement of our histology-based, AI-guided diagnostics tools such as RlapsRisk BC (lately MSIntuit CRC); which can be done through complex model understanding and data representation. As such, I have been working within R&D on self-supervised learning and released Phikon, Owkin’s foundation model for histology images.

Past experience in computer vision and medical research

After graduading, I spent two years in the Include team, the data wharehouse of Lille University Hospital (France) created in 2018 and authorized by the CNIL to reuse patients’ data for clinical and methodological research. I gained expertise in (bio)statistics and machine learning techniques (e.g. clustering, survival analysis, time series analysis) through diverse collaborations with researchers and phyicians. In the mean time, I started to specialize in computational pathology through a 2-year collaboration with Dr. Florence Renaud, pathologist, on the prediction of molecular subtypes in esogastric adenocarcinomas. Then, I joined the ENDOMIC (Inserm | INRIA) team as a data scientist. Here, I continued working on computational pathology but also on the analysis of immunofluorescence images through a 1-year collaboration with the immunology institute of Lille University Hospital (Pr. Sylvain Dubucquoi, Pr. Vincent Sobanski).