RARE: identification and tracking of cell populations of interest in predictive medicine

Identification and tracking of cell populations of interest in predictive medicine using static learning methods applied to flow cytometry data

CONTEXT
Flow cytometry data analysis software currently performs classic tasks such as identifying cell subpopulations and displaying results. Some solutions offer multi-parameter analysis, but they do not provide automated and reliable handling of the rare events occurring in numerous diseases.

For the most part, rare circulating cells are associated with cancer phenomena, whether within the scope of solid tumors or hematological malignancies. These cells may nonetheless also be associated with residual diseases in infectiology, adaptive immune responses, or in the context of vascular pathologies.

Depending on the treatment received by patients, or the nature of the cell population, these rare circulating cells may circulate within the organism with frequency lower than 10-5[MI1] . In order to prevent relapses, it is therefore necessary to identify these rare circulating cells and be able to analyze a larger number of events.

BENEFITS
The objective of the RARE project is to develop a tool to automatically identify cells of interest in some diseases by leveraging a specific data mining algorithm. These cells may also represent biological markers that are important in predictive medicine to track chronic conditions with heavy societal consequences, including vascular pathologies, oncology, infectiology, and onco-hematology (including acute myeloid leukemia, which will serve as a model to validate the tool).

APPLICATIONS
Vascular pathologies, oncology, infectiology, onco-hematology, detection of residual disease