MEDomicsLab

The MEDomicsLab research laboratory focuses on the development of predictive models with heterogeneous medical data.

ByProfessor
Martin Vallières.
Martin Vallières

to presentAssociate Professor,
Department of Oncology,
McGill University.

to Assistant Professor,
Department of Computer Science,
Université de Sherbrooke.

Principal Investigator

Martin
Vallières, PhD

Explore Martin’s profile Meet our team

We advance precision medicine by building open-source, clinically grounded AI tools and methods spanning medical imaging, multimodal predictive modeling, and privacy-aware collaboration. Open science. Open source.

TheMEDomicsEcosystem.

Scroll to learn more about our

Publications · News · Events

Featured
Publications

Come check out
our latest news

Visit all news
Conferences

Participation in Mila's Community of Practice on Digital Health

Thesis Defenses

Mariem Kallel successfully defended her thesis

Publications

Publication of the article “Predictive performance precision analysis in medicine: identification of low-confidence predictions at patient and profile levels (MED3pa I)”

Come to
our events.

Visit all events

Where next?

Come explore the lab’s projects, meet the team, or get in touch to discuss a possible collaboration.

Our research Meet the team Get in touch
Development of the MEDomics platformDevelopment of the MEDiml packagePredictive Performance Precision Analysis in Medicine (MED3pa)Predictive modeling based on multi-level graphical representations of multimodal healthcare dataOptimizing MEDomics platform: Enhancing Usability for Clinical AI AdoptionDevelopment of a methodological review on statistical inference and distributed learning in healthRandom forests in a distributed learning contextDevelopment of artificial intelligence techniques for the automated identification of electrical energy assetsRadiomics for the management of small renal massesMulti-task learning for image classification of renal tumorsGraph neural networks for predicting patient mortality within one year of hospital admissionResilient predictive models based on quantitative imaging to guide prostate cancer treatmentExploration of the complexity levels of radiomic characteristicsMachine learning strategies for the diagnostic and the prediction of late adverse effects related to childhood lymphoblastic leukaemia treatment.