Making precision medicine real

Bringing the right therapy to the right patient

What we do

A digital platform for personalized medicine for immuno-oncology and inflammatory dysfunctions.

For whom

Our platform serves all stakeholders wanting to use complementary diagnostics for improving patient outcomes.

Our approach

The platform addresses the scarcity of predictive and prognostic biomarkers, using strong clinical rationale, machine learning and statistical methods.

Meet the team

Dora Sabino


Ph.D. Cancer Research UK, Cambridge. 15 years of experience in research, translational & fundamental. Development of R&D for pharma & AI clients.

Leo Fillioux

Cofounding ML engineer

Masters machine learning, CentraleSupélec. 3+ years work experience in high-impact startups. Focus on machine learning applied to oncology.


Multimodal patient data.

Combining multiple sources of data allows us to explore biomarkers from combinations of modalities which would be impossible to explore by a human being.

Learn more

State of the art machine learning platform.

Using latest advances in the field of computer vision, deep learning, and machine learning, we perform a multimodal data analysis with various endpoints.

Learn more

Improved patient outcomes.

Our models allow us to better divide patients into groups, so as to better target patients who will receive a specific treatment. This leads to better patient outcomes for both predicted responders and non-responders.

Learn more



Detecting biomarkers of response to immunotherapies and other cancer therapies to give the right therapy to the right patient, improving patient outcome. Blazar's approach is to generate complementary diagnostics that are going to speed up and give accuracy to clinical decisions but also contribute to the discovery of new digital biomarkers with clinical and industrial partners.


Platform to help pathologists report on cancer types and analyze histopathology whole slide images with productivity tools (automatic segmentation of regions of interest, report generation, assisting in the annotations, …).

Want to partner with us?

Contact us