Grant: Master's Award: Canada Graduate Research Scholarships - Master's program
University of Toronto
$27,000.00
Precision Medicine-based Risk Assessment System for Patients with Concussion Using Machine Learning
FY 2025-26
Summary
A concussion is a temporary change in how the brain works. It can occur after sports, falls, motor-vehicle collisions, or everyday accidents. Most people recover within days or weeks, but some have headaches, dizziness, thinking difficulties, or light and noise sensitivity for months or longer. Doctors rely on symptom checklists and brief memory and balance tests, but these tools cannot reliably predict who will recover quickly and who may take longer. This uncertainty makes planning for daily life difficult. Recovery is complex and influenced by many factors, including thinking and memory skills, balance, mood, sleep, past health, and each person's unique demands. No single test captures all these areas, and there is no clinical tool that provides individualized recovery predictions.
This study uses data from over 3,000 people who experienced a concussion. All records are de-identified and stored securely on the Ontario Brain Institute's Brain-CODE platform. The dataset includes symptom scores, cognitive tests, gait and balance measures, eye-tracking results, mental-health information, and for one of the first times, detailed background and life-history factors. These data can help us understand concussion at a deeper and more individualized level and support personalized recovery predictions. All data are de-identified and analyzed with machine learning inside a secure environment. The study will develop models that combine many types of information to improve diagnostic accuracy and to predict whether a person will recover quickly or more slowly. The models will also identify the factors that shape recovery. The findings may support future tools, like a calculator, that help clinicians better understand recovery trajectories and guide planning for follow-up care.
Department
Program
Master's Award: Canada Graduate Research Scholarships - Master's program
Recipient
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database
As published
Id
572937
Name
Milne Orson B
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Precision Medicine-based Risk Assessment System for Patients with Concussion Using Machine Learning
Programtype2
Studentship Programs
Orgnameinp2
University of Toronto
Cihramount2
$27,000
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Research Scholarships - Master's program
Keyworddelim
Concussion; Diagnostic Modelling; External Validation; Gait And Oculomotor Biomarkers; Machine Learning; Mild Traumatic Brain Injury; Multimodal Data Integration; Neurocognitive Assessment; Precision Medicine; Prognostic Modelling
Abstract
A concussion is a temporary change in how the brain works. It can occur after sports, falls, motor-vehicle collisions, or everyday accidents. Most people recover within days or weeks, but some have headaches, dizziness, thinking difficulties, or light and noise sensitivity for months or longer. Doctors rely on symptom checklists and brief memory and balance tests, but these tools cannot reliably predict who will recover quickly and who may take longer. This uncertainty makes planning for daily life difficult. Recovery is complex and influenced by many factors, including thinking and memory skills, balance, mood, sleep, past health, and each person's unique demands. No single test captures all these areas, and there is no clinical tool that provides individualized recovery predictions.
This study uses data from over 3,000 people who experienced a concussion. All records are de-identified and stored securely on the Ontario Brain Institute's Brain-CODE platform. The dataset includes symptom scores, cognitive tests, gait and balance measures, eye-tracking results, mental-health information, and for one of the first times, detailed background and life-history factors. These data can help us understand concussion at a deeper and more individualized level and support personalized recovery predictions. All data are de-identified and analyzed with machine learning inside a secure environment. The study will develop models that combine many types of information to improve diagnostic accuracy and to predict whether a person will recover quickly or more slowly. The models will also identify the factors that shape recovery. The findings may support future tools, like a calculator, that help clinicians better understand recovery trajectories and guide planning for follow-up care.
Competitiondate
202512
Approvedterm2
1 yr 0 mth
Pinamesdelim
Milne, Orson B
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $27,000~ $0~