Decentralized Deep Radiomics: Scaling up the discovery of prognostic and predictive cancer imaging biomarkers from routine clinical data across a network of hospitals

Genetic advances over the past 10 years have led to the development of several targeted therapies for lung, breast and colon cancer. However, there are a number of factors that limit the optimal use of these innovations, including the high cost of the organizational process associated with molecular testing, and their late use in the patient's journey. Recently, the prospect of obtaining non-invasive, cost-effective and timely triggers for diagnostic & therapy has emerged from a discipline known as Radiomics. Radiomics leverages advances in artificial intelligence (AI) for the quantitative extraction of high-dimensional imaging features with promising predictive and prognostic indications. Nonetheless, a lack of evidence from large enough cohort validation, and the transdisciplinarity necessary for this process appears to be a methodological barrier spanning from the physics and treatment of medical imaging, the analysis of big data in health (clinical, biological, genetic and epigenetic), to the required expertise in machine & deep learning. The proposed project aims at operationalizing at a large scale Imagia’s clinical evidence ecosystem, called Evidens, to enable a sharp decrease in the cost of discovering predictive and prognostic cancer imaging biomarkers, by pursuing and applying fundamental advances in machine learning, data privacy, distributed computing and biostatistics.

Kun (Ryan) Ni;Marie St-Laurent
Superviseur universitaire: 
Yoshua Bengio;Caroline Reinhold;Aaron Courville;Jaron Chong
Partner University: