Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Master's Award: Canada Graduate Scholarships

Université Laval

$17,500.00

Molecular imaging to optimize the care of patients with advanced prostate cancer: connecting the dots between radiomics and transcriptomics using artificial intelligence

FY 2021-22

Summary
Prostate cancer (PCa) has the second leading cancer-specific mortality in Canadian men. At time of diagnosis, cancer's aggressiveness is determined to predict the risk of mortality. Molecular imaging can identify distinct features present in cancerous tissues and provide valuable information on their aggressiveness in real time. Radiomics is an innovative approach that extracts quantitative data from medical images. We therefore believe that by harnessing the potential of radiomics, we can produce a predictive smart tool to identify at time of diagnosis high-risk patients to optimize their care and improve their survival.Our goals are (1) to develop a radiomics-based model guided by artificial intelligence to predict relevant clinical outcomes, (2) to analyze the genetic expression of cancerous tissues with variable metabolism visualized on molecular imaging to understand their biological processes, and (3) to implement these data into the tool to increase its predictive accuracy of disease evolution.This is a retrospective study including 343 patients with high-grade PCa that are imaged using molecular imaging between 2011 and 2021 at our institution. Firstly, a radiomics-based model will be developed, and then tested and validated using two distinct sub-cohorts. This model will predict the following endpoints: lymph nodes invasion, time to biochemical recurrence after surgery, time to metastases, time to castration resistance, and cancer-specific mortality. PCa tissues and regional lymph nodes specimens obtained from patients after surgery are analyzed to identify biological processes associated with cancer metabolism. Radiomics and biological data will be processed in a machine learning algorithm to develop a tool predicting the endpoints with better accuracy.
Program
Master's Award: Canada Graduate Scholarships
Location
Quebec, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database

As published

Id
466805
Name
Touma Nawar
Region
Québec
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Molecular imaging to optimize the care of patients with advanced prostate cancer: connecting the dots between radiomics and transcriptomics using artificial intelligence
Programtype2
Studentship Programs
Orgnameinp2
Université Laval
Cihramount2
$17,500
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Scholarships
Keyworddelim
Artificial Intelligence; F-Fdg; Molecular Imaging; Pet/Ct; Predictive Model; Prostate Cancer; Radiomics; Transcriptomics
Abstract
Prostate cancer (PCa) has the second leading cancer-specific mortality in Canadian men. At time of diagnosis, cancer's aggressiveness is determined to predict the risk of mortality. Molecular imaging can identify distinct features present in cancerous tissues and provide valuable information on their aggressiveness in real time. Radiomics is an innovative approach that extracts quantitative data from medical images. We therefore believe that by harnessing the potential of radiomics, we can produce a predictive smart tool to identify at time of diagnosis high-risk patients to optimize their care and improve their survival.Our goals are (1) to develop a radiomics-based model guided by artificial intelligence to predict relevant clinical outcomes, (2) to analyze the genetic expression of cancerous tissues with variable metabolism visualized on molecular imaging to understand their biological processes, and (3) to implement these data into the tool to increase its predictive accuracy of disease evolution.This is a retrospective study including 343 patients with high-grade PCa that are imaged using molecular imaging between 2011 and 2021 at our institution. Firstly, a radiomics-based model will be developed, and then tested and validated using two distinct sub-cohorts. This model will predict the following endpoints: lymph nodes invasion, time to biochemical recurrence after surgery, time to metastases, time to castration resistance, and cancer-specific mortality. PCa tissues and regional lymph nodes specimens obtained from patients after surgery are analyzed to identify biological processes associated with cancer metabolism. Radiomics and biological data will be processed in a machine learning algorithm to develop a tool predicting the endpoints with better accuracy.
Competitiondate
202112
Approvedterm2
1 yr 0 mth
Pinamesdelim
Touma, Nawar
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $17,500~ $0~