New kids on the block: Improving the care of people of people with multiple sclerosis by introducing “Radiomics” en “Epomics”
New kids on the block: Improving the care of people of people with multiple sclerosis by introducing “Radiomics” en “Epomics”
(Hamza Khan - ongoing since September 2020)
Joint PhD together with UMaastricht
The intent of this doctoral research is to use machine learning to speed up the search for the optimal disease modifying therapy by using real-world data. The hypothesis is that with the help of machine learning, we can find features from “Radiomics” and “Epomics” that are sensitive to MS disease progression compared to the statistical models that only use classical, well-known features such as number of lesions for MRI and latency for evoked potential time series (EPTS).
Radiomics is a field of medical study that aims to extract a large number of quantitative features from medical images (here: MRI’s) using data characterization algorithms. Radiomics has the potential to uncover disease characteristics that are difficult to identify by human vision alone
Epomics refer to the features extracted from EPTS via machine learning techniques. It allows us to use the full information captured in the high-dimensional EPTS.
The PhD project focuses on following objectives:
- Objective 1: Investigate whether Radiomics biomarkers can support the identification of MRI lesions resulting in functional damage
- Objective 2: Investigate whether by adding Radiomics and Epomics features, we can improve predictions of long-term disability progression
- Objective 3: Investigate whether short-term changes in Radiomics and Epomics features predict long-term disability progression
Hamza is involved in following subprojects:
POC3 of the Flanders AI Research Project (2019-2023):
- Short-term differences in radiomic and/or evoked potential time series features predict long-term treatment effectiveness
- ML-based harmonization of MR images from MS patients as a preprocessing step in a pipeline for predicting long-term disability progression
MSDA:
- GDSI open data
- EHDEN - transforming the MS DataConnect dataset to OMOP-CDM
- MS DataConnect (and medEmotion)