Grant: Master's Award: Canada Graduate Scholarships
University of Calgary
$27,000.00
Predicting disease progression in people with secondary progressive multiple sclerosis using deep learning of brain MRI
FY 2024-25
Summary
Introduction:Secondary progressive multiple sclerosis (SPMS), a severe stage of MS, is marked by widespread inflammation and progressive neurological decline. Unlike relapsing-remitting MS, SPMS lacks effective tools for progression assessment. While MRI is critical in studying MS, manual interpretation is time-consuming and variable. Deep learning (DL) offers a solution by automating lesion detection and identifying subtle disease patterns.Hypothesis and Aims:This study hypothesizes that DL models can predict SPMS progression using MRI data, enabling biomarker discovery for personalized treatment. The aims are to (1) develop DL models for SPMS progression prediction (Year 1) and (2) interpret model results to identify key biomarkers (Year 2).Methods:Using MRI data from 120 SPMS patients in the CIMS study, preprocessed brain scans (T1-weighted, T2-weighted, and FLAIR) will train DL models such as ResNet50 and Vision Transformer. Models will predict changes in the Expanded Disability Status Scale (EDSS) over five years, validated through five-fold cross-validation. Feature importance will be assessed using Shapley Additive Explanations (SHAP), linking MRI features to clinical outcomes.Conclusion:This research aims to develop DL-based tools for SPMS progression prediction and biomarker identification, addressing gaps in SPMS research and enabling personalized treatment strategies.
Department
Program
Master's Award: Canada Graduate Scholarships
Recipient
Location
Alberta, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database
As published
Id
530047
Name
Salman Muhammad Saahim
Region
Alberta
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Predicting disease progression in people with secondary progressive multiple sclerosis using deep learning of brain MRI
Programtype2
Studentship Programs
Orgnameinp2
University of Calgary
Cihramount2
$27,000
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Scholarships
Keyworddelim
Artificial Intelligence; Brain; Deep Learning; Disease Progression; Imaging; Missing Data; Mri; Multiple Sclerosis; Nervous System; Transfer Learning
Abstract
Introduction:Secondary progressive multiple sclerosis (SPMS), a severe stage of MS, is marked by widespread inflammation and progressive neurological decline. Unlike relapsing-remitting MS, SPMS lacks effective tools for progression assessment. While MRI is critical in studying MS, manual interpretation is time-consuming and variable. Deep learning (DL) offers a solution by automating lesion detection and identifying subtle disease patterns.Hypothesis and Aims:This study hypothesizes that DL models can predict SPMS progression using MRI data, enabling biomarker discovery for personalized treatment. The aims are to (1) develop DL models for SPMS progression prediction (Year 1) and (2) interpret model results to identify key biomarkers (Year 2).Methods:Using MRI data from 120 SPMS patients in the CIMS study, preprocessed brain scans (T1-weighted, T2-weighted, and FLAIR) will train DL models such as ResNet50 and Vision Transformer. Models will predict changes in the Expanded Disability Status Scale (EDSS) over five years, validated through five-fold cross-validation. Feature importance will be assessed using Shapley Additive Explanations (SHAP), linking MRI features to clinical outcomes.Conclusion:This research aims to develop DL-based tools for SPMS progression prediction and biomarker identification, addressing gaps in SPMS research and enabling personalized treatment strategies.
Competitiondate
202412
Approvedterm2
1 yr 0 mth
Pinamesdelim
Salman, Muhammad Saahim
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $27,000~ $0~