Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Catalyst Grant: NWHRI: Innovation Fund - Biomedical Discovery Grants

Research Institute of the McGill University Health Centre

$150,000.00

Magnetic Resonance Imaging (MRI) and Artificial Intelligence (AI) can improve preoperative malignancy risk prediction of ovarian masses.

EF 2023-24

Résumé
Ovarian Cancer (OC) has a poor prognosis in the majority of patients & providing accurate & timely diagnosis can be life saving. Reducing unnecessary surgeries can provide access to those with a high likelihood of OC. As a patient, having an accurate diagnosis sooner would eliminate anxiety for those whose conditions are benign. Although ultrasound (US) plays an important role in the initial detection of OC, 5-25% of ovarian masses will remain indeterminate on US. Magnetic Resonance Imaging (MRI) is the modality of choice for characterizing ovarian masses that are indeterminate on US. The ovarian MRI risk stratification system (O-RADS MRI) uses both anatomic & functional imaging & assigns a risk score of cancer to an ovarian mass. The aim of the O-RADS MRI is to improve the accuracy & reproducibility of characterizing OC on MRI with the overarching goal to avoid unnecessary surgery in women with benign lesions & timely access to oncological surgery for those with high-risk lesions. This is critical in our resource strapped health care system to ensure optimal care for those patients that require it most. We will (a )validate the performance & inter-observer variability of the O-RADS MRI risk stratification on patients with ovarian lesions (b) develop novel MRI biomarkers to more accurately diagnose OC in the low risk (5%) O-RADS MRI 3 category & the intermediate risk (50%) O-RADS 4 category & (c) develop an artificial intelligence (AI) algorithm to predict benign vs. malignant lesions. The use of AI has great potential to effectively construct predictive models for analyzing complex multi-dimensional datasets. Unlike traditional computer-aided diagnostic support tools, AI algorithms can learn, adapt, & improve with experience. With the advent of increasingly complex targeted therapies, AI-supported approaches represent a new horizon for the development of clinical decision support tools using non-invasive MRI-based biomarkers for guiding precision care for women.
Programme
Catalyst Grant: NWHRI: Innovation Fund - Biomedical Discovery Grants
Emplacement
Quebec, Canada
Type
Grant
Source
CIHR Funding Decisions Database

Tel que publié

Id
486885
Name
Reinhold Caroline
Region
Québec
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Cancer Research
Theme2
Clinical
Prcname2
NWHRI: Innovation Fund - Biomedical Discovery Grants
Projecttitle
Magnetic Resonance Imaging (MRI) and Artificial Intelligence (AI) can improve preoperative malignancy risk prediction of ovarian masses.
Programtype2
Operating Grants
Orgnameinp2
Research Institute of the McGill University Health Centre
Cihramount2
$150,000
Cihrequipment2
$0
Programname2
Catalyst Grant: NWHRI: Innovation Fund - Biomedical Discovery Grants
Keyworddelim
Adnexal Masses; Artificial Intelligence; Biomarkers; Detection; Histopathology; Magnetic Resonance Imaging; Ovarian-Adnexal Imaging Reporting System(O-Rads); Ultrasound
Abstract
Ovarian Cancer (OC) has a poor prognosis in the majority of patients & providing accurate & timely diagnosis can be life saving. Reducing unnecessary surgeries can provide access to those with a high likelihood of OC. As a patient, having an accurate diagnosis sooner would eliminate anxiety for those whose conditions are benign. Although ultrasound (US) plays an important role in the initial detection of OC, 5-25% of ovarian masses will remain indeterminate on US. Magnetic Resonance Imaging (MRI) is the modality of choice for characterizing ovarian masses that are indeterminate on US. The ovarian MRI risk stratification system (O-RADS MRI) uses both anatomic & functional imaging & assigns a risk score of cancer to an ovarian mass. The aim of the O-RADS MRI is to improve the accuracy & reproducibility of characterizing OC on MRI with the overarching goal to avoid unnecessary surgery in women with benign lesions & timely access to oncological surgery for those with high-risk lesions. This is critical in our resource strapped health care system to ensure optimal care for those patients that require it most. We will (a )validate the performance & inter-observer variability of the O-RADS MRI risk stratification on patients with ovarian lesions (b) develop novel MRI biomarkers to more accurately diagnose OC in the low risk (5%) O-RADS MRI 3 category & the intermediate risk (50%) O-RADS 4 category & (c) develop an artificial intelligence (AI) algorithm to predict benign vs. malignant lesions. The use of AI has great potential to effectively construct predictive models for analyzing complex multi-dimensional datasets. Unlike traditional computer-aided diagnostic support tools, AI algorithms can learn, adapt, & improve with experience. With the advent of increasingly complex targeted therapies, AI-supported approaches represent a new horizon for the development of clinical decision support tools using non-invasive MRI-based biomarkers for guiding precision care for women.
Competitiondate
202307
Approvedterm2
1 yr 0 mth
Pinamesdelim
Reinhold, Caroline
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $150,000~ $0~