Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Project Grant

McMaster University

$110,926.00

Phenotyping heart failure with preserved ejection fraction

FY 2023-24

Summary
Heart failure (HF) occurs when the heart cannot function properly to meet the demands of the body, and once it occurs, the average survival is only 5 years. HF with preserved ejection fraction (HFpEF) is a complex syndrome that can result from multiple different risk factors, biological pathways, and changes to the heart's structure or function. There is a need to better understand how these features relate to one another. Machine learning (ML) is a branch of artificial intelligence that is well suited to finding patterns in complex data. We will use unsupervised ML methods to identify distinct groups of patients with shared clinical characteristics (or 'phenotypes') in a study of 4686 participants with HFpEF, then determine how these groups differ in terms of clinical features and prognosis. In a subset of these participants, we have collected blood samples and measured levels of over 500 protein markers involved in different biological pathways that may be related to HF. Using these data, we will identify markers of disease progression in HFpEF, and determine whether these markers differ among HFpEF phenotypes. Results will improve on the current understanding of HFpEF, and may inform more targeted approaches to the treatment of this complex form of HF.
Program
Project Grant
Location
Ontario, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database

As published

Id
501071
Name
Joseph Philip G
Region
Ontario
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Circulatory and Respiratory Health
Theme2
Clinical
Prcname2
Clinical Investigation - D: Cardiovascular Systems
Projecttitle
Phenotyping heart failure with preserved ejection fraction
Programtype2
Operating Grants
Orgnameinp2
McMaster University
Cihramount2
$110,926
Cihrequipment2
$0
Programname2
Project Grant
Keyworddelim
Heart Failure; Machine Learning; Phenotype; Proteomics
Abstract
Heart failure (HF) occurs when the heart cannot function properly to meet the demands of the body, and once it occurs, the average survival is only 5 years. HF with preserved ejection fraction (HFpEF) is a complex syndrome that can result from multiple different risk factors, biological pathways, and changes to the heart's structure or function. There is a need to better understand how these features relate to one another. Machine learning (ML) is a branch of artificial intelligence that is well suited to finding patterns in complex data. We will use unsupervised ML methods to identify distinct groups of patients with shared clinical characteristics (or 'phenotypes') in a study of 4686 participants with HFpEF, then determine how these groups differ in terms of clinical features and prognosis. In a subset of these participants, we have collected blood samples and measured levels of over 500 protein markers involved in different biological pathways that may be related to HF. Using these data, we will identify markers of disease progression in HFpEF, and determine whether these markers differ among HFpEF phenotypes. Results will improve on the current understanding of HFpEF, and may inform more targeted approaches to the treatment of this complex form of HF.
Competitiondate
202403
Approvedterm2
1 yr 0 mth
Pinamesdelim
Joseph, Philip G
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $110,926~ $0~