Key result
CatBoost machine learning achieves 93% accuracy in distinguishing CVD patients from healthy controls.
Why the study?
CVD is the leading cause of mortality in Middle Eastern countries including Qatar, but no comprehensive study had identified Qatar-specific CVD risk factors.
Does a machine learning model using multimodal biomedical measurements improve the identification of cardiovascular disease and its risk factors compared to traditional risk factors in a Qatari cohort?
Population
Healthy individuals and people with CVD from the Qatar Biobank
Comparison
People with CVD vs healthy individuals
Design
Case-control study
Authors
Loading...
May support ML-based CVD detection in Middle Eastern cohorts; leaves open clinical utility pending prospective validation.
Case-Control (n=500)
No
Does a machine learning model using multimodal biomedical measurements improve the identification of cardiovascular disease and its risk factors compared to traditional risk factors in a Qatari cohort?
Effect estimate: null (95% CI null)
p-value: p=<0.001
A machine learning model incorporating multimodal clinical and bioimpedance data accurately identified CVD patients in Qatar and revealed novel risk factors such as hypercoagulability and renal disorder markers.
Al-Absi et al. (2021) conducted a case-control in Cardiovascular Disease (CVD) (n=500). CatBoost machine learning model vs. Control group of healthy individuals was evaluated on Classification of CVD group versus control group (null, 95% CI null, p=<0.001). The CatBoost machine learning model achieved an accuracy of 93% in distinguishing patients with cardiovascular disease from healthy controls in a Qatari population cohort.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: