Grant: DRA : Canada Graduate Research Scholarship - Doctoral Program (CGRS D)
University of Calgary
$120,000.00
Machine learning in MRI to predict meningioma methylation subclassification
FY 2025-26
Summary
Meningioma is the most common primary brain tumor in adults. The standard and readily available method for diagnosing meningioma is to examine tumor cells under a microscope. However, recent studies indicate that this approach is not entirely accurate, as approximately one in four tumors labeled as benign actually exhibit more aggressive behavior with higher recurrence and growth rates. Incorporating molecular markers could improve diagnosis, but genetic and epigenetic analyses are costly and challenging for resource-limited centers, even in developed countries. Preoperative tools to evaluate tumor aggressiveness can aid in surgical planning and post-surgery management. Machine learning shows promise as a tool for this purpose. Magnetic resonance imaging (MRI) produces a wealth of radiological data, much of which is undetectable to the naked eye, but artificial intelligence can extract thousands of features from large MRI datasets. These features can be linked to clinical data, tumor grade, molecular genetics, surgical anatomy, and recurrence risk. We aim to develop a machine learning model that extracts features from MRI scans to predict molecular changes in meningioma tumors. This project aims to create an AI tool that uses MRI data to classify meningioma molecular subtypes. We believe MRI features can sufficiently predict tumor biology and recurrence risk. This is the first study to explore machine learning for predicting meningioma methylation subclasses. The findings could significantly impact surgical planning and enable personalized treatment strategies for patients.
Department
Program
DRA : Canada Graduate Research Scholarship - Doctoral Program (CGRS D)
Recipient
Location
Alberta, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database
As published
Id
559827
Name
Karmur Brij S
Region
Alberta
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Cancer Research
Theme2
Clinical
Prcname2
DT1 (Biomedical and/or Clinical Research) CGRS Doctoral
Projecttitle
Machine learning in MRI to predict meningioma methylation subclassification
Programtype2
Studentship Programs
Orgnameinp2
University of Calgary
Cihramount2
$120,000
Cihrequipment2
$0
Programname2
DRA : Canada Graduate Research Scholarship - Doctoral Program (CGRS D)
Keyworddelim
Ai; Artificial Intelligence; Cnn; Convolutional Neural Network; Deep Learning; Machine Learning; Meningioma; Methylation; Mri
Abstract
Meningioma is the most common primary brain tumor in adults. The standard and readily available method for diagnosing meningioma is to examine tumor cells under a microscope. However, recent studies indicate that this approach is not entirely accurate, as approximately one in four tumors labeled as benign actually exhibit more aggressive behavior with higher recurrence and growth rates. Incorporating molecular markers could improve diagnosis, but genetic and epigenetic analyses are costly and challenging for resource-limited centers, even in developed countries. Preoperative tools to evaluate tumor aggressiveness can aid in surgical planning and post-surgery management. Machine learning shows promise as a tool for this purpose. Magnetic resonance imaging (MRI) produces a wealth of radiological data, much of which is undetectable to the naked eye, but artificial intelligence can extract thousands of features from large MRI datasets. These features can be linked to clinical data, tumor grade, molecular genetics, surgical anatomy, and recurrence risk. We aim to develop a machine learning model that extracts features from MRI scans to predict molecular changes in meningioma tumors. This project aims to create an AI tool that uses MRI data to classify meningioma molecular subtypes. We believe MRI features can sufficiently predict tumor biology and recurrence risk. This is the first study to explore machine learning for predicting meningioma methylation subclasses. The findings could significantly impact surgical planning and enable personalized treatment strategies for patients.
Competitiondate
202511
Approvedterm2
3 yrs 0 mth
Pinamesdelim
Karmur, Brij S
Approvedterm
3 ans 0 mois
Cihrcontribution2
~ $120,000~ $0~