Longitudinal Integrative Models for Online Relapse Detection: LIMORD

The goal of LIMORD is to propose a novel statistical tool allowing patient classification, earlier relapse detection and better prognosis estimation in order to move forward into personalized medicine in Multiple Myeloma.

It has several key components:

  • Development of a probabilistic model based on the formalism of Partially Observable Markov Decision Processes (POMDP) to integrate marker follow-up data and multiple sequencing datasets to detect patient relapse and accurately assign patients to relapse types
  • Development of machine-learning based tools to identify RNA modifications from Direct RNA sequencing data, development of a statistical tool to evaluate the degree of RNA fragmentation in patient samples, and prediction of response to RNA-modification targeted drugs
  • Development of a software to integrate the project results and deliver it to clinicians