Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Project Grant

McGill University

$271,576.00

A Bayesian deep neural network for early diagnosis of oral cancers

EF 2019-20

Résumé
More than 300,000 individuals diagnosed with oral cancer every year worldwide. Approximately half of all oral cancers are diagnosed at an advanced stage disease. The advanced stage at diagnosis, compared to early-stage, is associated with a lower survival rate (<50%) and higher morbidity and healthcare cost. Early diagnosis and treatment can significantly improve the survival probabilities and quality of life among these patients. However, limited access to a specialist is a major contributor to delayed diagnosis. An abnormal lesion is one of the early signs of oral cancer. Surprisingly, many primary care practitioners find it challenging to diagnose cancerous oral lesions due to limited training and a wide variety of lesions that appear clinically similar. This research project will develop a machine learning-based tool that can help doctors to diagnose cancerous oral lesions using images. We will use a large data set of images of oral lesions compiled from different parts of Canada and abroad, to develop our tool. Further, we will compare the accuracy of the tool with the accuracy of oral cancer experts across Canada.
Programme
Project Grant
Bénéficiaire
Emplacement
Quebec, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database

Tel que publié

Id
421110
Name
Madathil Sreenath A
Region
Qu&eacute;bec
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Musculoskeletal Health and Arthritis
Theme2
Clinical
Prcname2
Health Services Evaluation &amp; Interventions Research 2
Projecttitle
A Bayesian deep neural network for early diagnosis of oral cancers
Programtype2
Operating Grants
Orgnameinp2
McGill University
Cihramount2
$271,576
Cihrequipment2
$0
Programname2
Project Grant
Keyworddelim
Clinical Prediction Model; Diagnostic Algorithm; Informative Prior Ditributions
Abstract
More than 300,000 individuals diagnosed with oral cancer every year worldwide. Approximately half of all oral cancers are diagnosed at an advanced stage disease. The advanced stage at diagnosis, compared to early-stage, is associated with a lower survival rate (<50%) and higher morbidity and healthcare cost. Early diagnosis and treatment can significantly improve the survival probabilities and quality of life among these patients. However, limited access to a specialist is a major contributor to delayed diagnosis. An abnormal lesion is one of the early signs of oral cancer. Surprisingly, many primary care practitioners find it challenging to diagnose cancerous oral lesions due to limited training and a wide variety of lesions that appear clinically similar. This research project will develop a machine learning-based tool that can help doctors to diagnose cancerous oral lesions using images. We will use a large data set of images of oral lesions compiled from different parts of Canada and abroad, to develop our tool. Further, we will compare the accuracy of the tool with the accuracy of oral cancer experts across Canada.
Competitiondate
202003
Approvedterm2
3 yrs 0 mth
Pinamesdelim
Madathil, Sreenath A; Tamimi, Faleh
Approvedterm
3 ans 0 mois
Cihrcontribution2
~ $271,576~ $0~