Multimodal digital biomarkers for the identification of companion diagnostic tools in cardiovascular diseases.

Challenge: The low efficacy of treatment is partially due to the heterogeneity of disease and population. Individuals do not respond the same way to treatment and can have very different clinical trajectories that are difficult to predict. In order to improve treatment efficacy, precision medicine approaches need to be considered. A refined selection of the population can help to enrich the investigated population of a clinical trial increasing the absolute effects of the trial. Therefore, a companion diagnostic can assist in identifying individuals that (1) are at increased risk of cardiovascular events, (2) are likely to benefit from treatment.

Solution: Using Artificial Intelligence (AI)-based predictive approaches and proprietary platform, the team intend to: (1) identify subjects who are unlikely, (2) identify subjects who are likely, if not treated, to develop a severe cardiovascular event in the near future (next 2 years), and (3) identify a subgroup of individuals who are more likely to benefit from treatment. Taking into account multiple modalities can provide valuable information and the use of machine learning approaches is very well adapted to these kinds of complex tasks where we need to take into account the relationship and interaction of multiple data points of a subject to predict its response to treatment.

Expected achievements/Impact: At the completion of the project, it is intended that a prognostic enrichment tool will be developed, on the basis of AI and multimodal data, that will enable to predict subpopulations presenting a higher risk of cardiovascular events. Deployed in the context of clinical trials, this tool will help refine patients recruitment, increasing the absolute effects of the trial. Led by Perceiv Research, this project is a great example of the applicability of AI-based predictive approaches toward precision medicine. The success of the project will bring visibility to Perceiv Research which also plans to commercialize its digital biomarkers through its platform to pharmaceutical sponsors, jointly with the MHI. DalCor Pharma will also benefit from the project since the tools, used as companion diagnostic and enrichment strategy, could be applied for its own clinical trials.

Principal Investigator:

Christian Dansereau
Perceiv Research Inc.


Jean-Claude Tardif 
Institut de Cardiologie de Montréal


Ongoing Project
$570,000 / 1.5 years


Supported by CQDM through:

And by co-funding partners

• DalCor Pharmaceuticals Canada Inc.