Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Master's Award: Canada Graduate Scholarships

McGill University

$17,500.00

Investigating trajectories of comorbidities among adult congenital heart disease populations: A pan-Canadian longitudinal population study

EF 2022-23

Résumé
The population of adults with Congenital Heart Disease (CHD) has substantially increased. As these patients get older, there is a higher tendency of developing co-occurring diseases (known as comorbidities). As these acquired comorbidities greatly impact the health outcomes of the CHD population, there is a need to understand how comorbidities evolve over the lifespan of patients with CHD (known as disease trajectories). However, we lack an effective method to capture the temporal patterns between specific diseases. To gain a greater understanding of the differences and similarities in disease trajectories with respect to several health outcomes, we propose: 1) to pinpoint the most common medical characteristics to categorize CHD patients into different subgroups at different ages; 2) to identify trajectories of comorbidities across CHD patient subgroups. The proposed work will examine the 138,000 patients included in the Quebec CHD database of 35 years of follow-up (1983–2017) and the 361,000 patients in the Canadian Institute of Health Information Discharge Abstract Database of 24 years’ hospitalization records (1994–2017). The proposed project will use seed-guided dynamic topic modeling to categorize CHD patients into latent groups. Dynamic topic modeling allows us to identify hidden temporal patterns between medical visits and diagnoses, which are used to group together patients with similar disease trajectories. These subgroups can in turn serve to predict the time of diagnoses of patients’ comorbidities as well as how they progress over time. Our study will be one of the first applications of machine learning methods to predict trajectories of comorbidities among CHD patients and will dramatically advance our understanding of how the disease progressed over time.
Programme
Master's Award: Canada Graduate Scholarships
Bénéficiaire
Emplacement
Quebec, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
CIHR Funding Decisions Database

Tel que publié

Id
498210
Name
Li Chao
Region
Québec
Country
Canada
Orgtype
Institution Paid; Research Institution
Prcname2
Special Cases - Awards Programs
Projecttitle
Investigating trajectories of comorbidities among adult congenital heart disease populations: A pan-Canadian longitudinal population study
Programtype2
Studentship Programs
Orgnameinp2
McGill University
Cihramount2
$17,500
Cihrequipment2
$0
Programname2
Master's Award: Canada Graduate Scholarships
Keyworddelim
Comorbidities; Congenital Heart Disease; Diabetes; Disease Trajectories; Machine Learning
Abstract
The population of adults with Congenital Heart Disease (CHD) has substantially increased. As these patients get older, there is a higher tendency of developing co-occurring diseases (known as comorbidities). As these acquired comorbidities greatly impact the health outcomes of the CHD population, there is a need to understand how comorbidities evolve over the lifespan of patients with CHD (known as disease trajectories). However, we lack an effective method to capture the temporal patterns between specific diseases. To gain a greater understanding of the differences and similarities in disease trajectories with respect to several health outcomes, we propose: 1) to pinpoint the most common medical characteristics to categorize CHD patients into different subgroups at different ages; 2) to identify trajectories of comorbidities across CHD patient subgroups. The proposed work will examine the 138,000 patients included in the Quebec CHD database of 35 years of follow-up (1983–2017) and the 361,000 patients in the Canadian Institute of Health Information Discharge Abstract Database of 24 years’ hospitalization records (1994–2017). The proposed project will use seed-guided dynamic topic modeling to categorize CHD patients into latent groups. Dynamic topic modeling allows us to identify hidden temporal patterns between medical visits and diagnoses, which are used to group together patients with similar disease trajectories. These subgroups can in turn serve to predict the time of diagnoses of patients’ comorbidities as well as how they progress over time. Our study will be one of the first applications of machine learning methods to predict trajectories of comorbidities among CHD patients and will dramatically advance our understanding of how the disease progressed over time.
Competitiondate
202212
Approvedterm2
1 yr 0 mth
Pinamesdelim
Li, Chao
Approvedterm
1 an 0 mois
Cihrcontribution2
~ $17,500~ $0~