Information détaillée
Avis
L'information est fournie dans la langue dans laquelle le ou la bénéficiaire l'a présentée.
- Titre du projet :
- Prostate cancer risk stratification for active surveillance using computer-aided analysis of histopathology and transcriptomics
- Chercheur principal :
- Bashashati, Ali; Black, Peter C; Goldenberg, Larry; Salcudean, Septimiu E
- Co-chercheurs :
- Fazli, Ladan; Gleave, Martin E; Macaulay, Calum E
- Directeur(s) de recherche :
- S/O
- Établisssement payé :
- University of British Columbia
- Établissement de recherche :
- University of British Columbia
- Département :
- Biomedical Engineering
- Programme :
- Subvention Projet
- Concours (année/mois) :
- 202010
- CEP désigné :
- Physique médicale et imagerie
- Institut principal :
- Cancer
- Thème principal :
- Recherche biomédicale
- Durée (année/mois) :
- 5 ans 0 mois
- Contribution des IRSC :
- Donateurs :
- Montant :
- 906 525$
- Équipement :
- 0$
- Contribution du partenaire externe :
- Nom du partenaire :
- S/O
- Montant :
- S/O
- Équipement :
- S/O
- Partenaire du candidat à l'externe :
- Nom du partenaire :
- S/O
- Montant :
- S/O
- Équipement :
- S/O
- Partenaire externe (en nature) :
- Nom du partenaire :
- S/O
- Montant :
- S/O
- Équipement :
- S/O
- Mots clés :
- Computational Pathology; Digital Pathology; Image Analysis; Machine Learning; Prostate Cancer; Transcriptomics
- Résumé :
- Prostate cancer remains the 2nd most common cause of cancer-related death in North American men. There are different risk categories; many that are diagnosed at an early stage will either become more dangerous over time, or will never impact on a man's longevity or general health. These latter cases can be managed with active surveillance, which avoids overtreatment and associated negative effects on quality of life. Current risk stratification tools, however, do not allow us to distinguish reliably which patients are best-suited for active surveillance and which are likely to develop higher-risk cancer over time. The artificial intelligence (AI) field is advancing at a staggering rate, accelerated by modern hardware and software. Across the board, healthcare is characterized by a tremendous proliferation of data, including huge amounts generated by digitizing pathology slides. A fundamental question is whether the computer can identify features in a biopsy that cannot be recognized by the human brain? We will deploy AI-based techniques to analyze digitized tumour tissue images and genetic markers in patients undergoing active surveillance. Our goal is to discover new pathologic and genetic markers that can more accurately stratify risk of progression in newly diagnosed prostate cancer so that a greater number can be monitored safely with active surveillance, including a subset that would be considered intermediate risk by current criteria, while higher risk cases can be selected for definitive treatment. Novel biomarkers will be extremely valuable to patients, caregivers and physicians in deciding on best treatment. This could dramatically decrease the risks of undesirable and unexpected cancer progression, and anxieties related to being on a cancer surveillance program. Further, the intensity of invasive and costly sequential prostate biopsies can be reduced in low-risk patients, leading to a better quality of life and decreased health care utilization.
- Version :
- 20250311.1