Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Master's Award: Canada Graduate Research Scholarships - Master's program

University of Toronto

$27,000.00

Enhancing Trauma Care Through Intelligent Surgical Volume Forecasting and Resource Optimization

EF 2025-26

Résumé
Every day in Canada, 9-14% of planned surgeries are cancelled at the last minute due to unforeseen emergency trauma cases that require immediate surgery. Current forecasting methods achieve only 42% accuracy, barely better than guessing, because they ignore crucial factors like weather, traffic patterns, and seasonal trends that influence when injuries occur. This creates impossible choices for hospitals: overstaff and waste resources or understaff and force cancellations while trauma patients wait for life-saving care. Our initial work has achieved ~80% accuracy in predicting orthopaedic surgical trauma volumes using traditional machine learning methods, but a different approach is needed to allow generalizability and improve temporal pattern understanding. This research pioneers the use of time series foundation models (TSFMs), a paradigm-shifting artificial intelligence (AI) technique, for surgical trauma volume prediction. TSFMs such as TimeGPT and Chronos are pre-trained on billions of data, excel at capturing long-range patterns, and can work well even with limited institution-specific data. The models will be trained using a database of 8.7 million trauma patient records, then fine-tuned with local data (including weather, traffic, and population). As AI can inadvertently worsen existing disparities, equity analysis (considering age, sex, gender, race/ethnicity) will be integral to our approach. Accurate forecasting means faster trauma care, better staff planning, fewer surgery cancellations, and more efficient operating room use. Leveraging models trained on massive multi-institutional databases can democratize healthcare AI across Canada, transforming surgical planning from reactive scrambling to proactive preparation that serves all patients more equitably.
Programme
Master's Award: Canada Graduate Research Scholarships - Master's program
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database

Tel que publié

Id
573011
Name
Ravindran Dharsan
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Enhancing Trauma Care Through Intelligent Surgical Volume Forecasting and Resource Optimization
Programtype2
Studentship Programs
Orgnameinp2
University of Toronto
Cihramount2
$27,000
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Research Scholarships - Master's program
Keyworddelim
Artificial Intelligence; Foundation Models; Health Administration; Healthcare Delivery; Machine Learning; Statistical Analysis; Surgery; Time Series; Trauma; Trend Forecasting
Abstract
Every day in Canada, 9-14% of planned surgeries are cancelled at the last minute due to unforeseen emergency trauma cases that require immediate surgery. Current forecasting methods achieve only 42% accuracy, barely better than guessing, because they ignore crucial factors like weather, traffic patterns, and seasonal trends that influence when injuries occur. This creates impossible choices for hospitals: overstaff and waste resources or understaff and force cancellations while trauma patients wait for life-saving care. Our initial work has achieved ~80% accuracy in predicting orthopaedic surgical trauma volumes using traditional machine learning methods, but a different approach is needed to allow generalizability and improve temporal pattern understanding. This research pioneers the use of time series foundation models (TSFMs), a paradigm-shifting artificial intelligence (AI) technique, for surgical trauma volume prediction. TSFMs such as TimeGPT and Chronos are pre-trained on billions of data, excel at capturing long-range patterns, and can work well even with limited institution-specific data. The models will be trained using a database of 8.7 million trauma patient records, then fine-tuned with local data (including weather, traffic, and population). As AI can inadvertently worsen existing disparities, equity analysis (considering age, sex, gender, race/ethnicity) will be integral to our approach. Accurate forecasting means faster trauma care, better staff planning, fewer surgery cancellations, and more efficient operating room use. Leveraging models trained on massive multi-institutional databases can democratize healthcare AI across Canada, transforming surgical planning from reactive scrambling to proactive preparation that serves all patients more equitably.
Competitiondate
202512
Approvedterm2
1 yr 0 mth
Pinamesdelim
Ravindran, Dharsan
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $27,000~ $0~