Prediction of heart failure with interpretable deep learning
Description: Heart failure (HF) is the most common cause of hospitalization for Ontarians over the age of 65, and is ultimately fatal for 50% of patients within 5 years of diagnosis. This is only expected to worsen in coming years due to Ontario’s aging population, and heretofore unknown fallout from the COVID-19 pandemic, both direct (changes in cardiac function from SARS-CoV-2 infection) and indirect (postponed medical treatments, surgeries, etc.) Early diagnosis is crucial to improve survivability and quality of life with HF, but current tools are insufficient for a timely response. With the influx of routinely collected administrative and clinical healthcare data, more information is available than ever before to inform clinical decision making, but a large team of clinicians working in tandem is typically required for each patient to benefit. We propose the use of powerful artificial intelligence (AI) tools to make use of these healthcare databases and develop an early-detection system for patients at high risk of HF. Through AI-assisted early detection of HF, we hope to not only provide the right healthcare at the right time for patients diagnosed with HF, but also mitigate the estimated $2.8 billion Canada will spend annually on HF care by 2030.
Amount: CA$140,000
Funding type: grant
Funder scheme: Early Researcher Awards
Start date: 2023-12-03
Investigator: Bo Wang
Institution: University of Toronto
Funded outputs count: 0
Funder award ID: ER22-17-202
OpenAlex ID: G4557413485