Grant: Master's Award: Canada Graduate Scholarships
University of Toronto
$17,500.00
Development of a computational pathology toolbox for advanced cancer diagnostics using artificial intelligence
EF 2020-21
Résumé
In 2020, approximately 225,000 Canadians are expected to be diagnosed with cancer. In each case, experienced pathologists analyze tumour slides and molecular data to reach a final diagnosis. Recently, advances in artificial intelligence (AI), specifically deep learning, have allowed AI-based programs to perform at an expert-level in medical image analyses, including classification of tumour slides. Most studies to date, however, have mainly shown proof-of-concepts rather than developing tangible platforms for widespread adoption by the research and medical communities. This project aims to address this issue by developing an online repository of computational tools that can begin refining many traditionally manual tasks in pathology. Ultimately, we aim to integrate several deep learning classifiers on this easily accessible website to generate pathologist-like reports. To do this, we will first develop a classifier that classifies common brain tumors based on their size and shape. Then we will develop molecular classifiers to subclassify the common brain tumor types and resolve challenging cases based on the presence of molecular markers. Both types of classifiers will be developed by training convolutional neural networks (CNNs) using images of different brain tumours. The tools will also be validated using test images for diagnosis accuracy. By the completion of this award, we will have developed the first AI-driven interface that integrates both molecular and histomorphologic information to automatically generate standardized expert-level neuropathology reports. By making these decision support tools widely available, this project is expected to accelerate the eventual routine adoption of deep learning in cancer research and clinical care.
Programme
Master's Award: Canada Graduate Scholarships
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database
Tel que publié
Id
449672
Name
Lee Michael K
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Development of a computational pathology toolbox for advanced cancer diagnostics using artificial intelligence
Programtype2
Studentship Programs
Orgnameinp2
University of Toronto
Cihramount2
$17,500
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Scholarships
Keyworddelim
Artificial Intelligence; Automation; Computational Pathology; Computer Vision; Decision Support; Deep Learning; Diagnosis; Image Analysis; Immunohistochemistry; Prognosis
Abstract
In 2020, approximately 225,000 Canadians are expected to be diagnosed with cancer. In each case, experienced pathologists analyze tumour slides and molecular data to reach a final diagnosis. Recently, advances in artificial intelligence (AI), specifically deep learning, have allowed AI-based programs to perform at an expert-level in medical image analyses, including classification of tumour slides. Most studies to date, however, have mainly shown proof-of-concepts rather than developing tangible platforms for widespread adoption by the research and medical communities. This project aims to address this issue by developing an online repository of computational tools that can begin refining many traditionally manual tasks in pathology. Ultimately, we aim to integrate several deep learning classifiers on this easily accessible website to generate pathologist-like reports. To do this, we will first develop a classifier that classifies common brain tumors based on their size and shape. Then we will develop molecular classifiers to subclassify the common brain tumor types and resolve challenging cases based on the presence of molecular markers. Both types of classifiers will be developed by training convolutional neural networks (CNNs) using images of different brain tumours. The tools will also be validated using test images for diagnosis accuracy. By the completion of this award, we will have developed the first AI-driven interface that integrates both molecular and histomorphologic information to automatically generate standardized expert-level neuropathology reports. By making these decision support tools widely available, this project is expected to accelerate the eventual routine adoption of deep learning in cancer research and clinical care.
Competitiondate
202012
Approvedterm2
1 yr 0 mth
Pinamesdelim
Lee, Michael K
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $17,500~ $0~