Improving the Detection of Potential Cases of Familial Hypercholesterolemia: Could Machine Learning Be Part of the Solution?

Aug 5, 2026

Upload by Ivy

Familial hypercholesterolemia (FH) is a highly prevalent but substantially underdiagnosed monogenic disorder that drives premature atherosclerotic cardiovascular events. This study evaluates whether machine learning models can outperform established clinical diagnostic criteria and UK-recommended screening criteria in identifying individuals likely to carry FH-causing genetic variants. Using UK Biobank participants with whole exome sequencing, the proposed stacking ensemble achieves the best predictive performance and reduces the number needed to screen versus baseline tools, supporting scalable implementation in electronic health records for prioritizing genetic confirmation.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Background

Methods and Results

Predictive modeling and evaluation

Conclusions

Clinical Perspective

Downloaded from [http://ahajournals.org by on June](http://ahajournals.org by on June) 19, 2024 Journal of the American Heart Association ORIGINAL RESEARCH | Improving the Detection of Potential Cases of Familial Hypercholesterolemia: Could Machine Learning Be Part of the Solution?
Christophe A. T. Stevens , MSc; Antonio J. Vallejo-Vaz , PhD; Joana R. Chora , PhD; Fotis Barkas , PhD; Julia Brandts , MD; Alireza Mahani, PhD; Leila Abar, PhD; Mansour T. A. Sharabiani, PhD*;
Kausik K. Ray , FMedSci*
BACKGROUND: Familial hypercholesterolemia (FH), while highly prevalent, is a significantly underdiagnosed monogenic disorder. Improved detection could reduce the large number of cardiovascular events attributable to poor case finding. We aimed to assess whether machine learning algorithms outperform clinical diagnostic criteria (signs, history, and biomarkers) and the recommended screening criteria in the United Kingdom in identifying individuals with FH-causing variants, presenting a scalable screening criteria for general populations.
METHODS AND RESULTS: Analysis included UK Biobank participants with whole exome sequencing, classifying them as having FH when (likely) pathogenic variants were detected in their LDLR, APOB, or PCSK9 genes. Data were stratified into 3 data sets for (1) feature importance analysis; (2) deriving state-of-the-art statistical and machine learning models; (3) evaluating models’predictive performance against clinical diagnostic and screening criteria: Dutch Lipid Clinic Network, Simon Broome, Make Early Diagnosis to Prevent Early Death, and Familial Case Ascertainment Tool. One thousand and three of 454 710 participants were classified as having FH. A Stacking Ensemble model yielded the best predictive performance (sensitivity, 74.93%; precision, 0.61%; accuracy, 72.80%, area under the receiver operating characteristic curve, 79.12%) and outperformed clinical diagnostic criteria and the recommended screening criteria in identifying FH variant carriers within the validation data set (figures for Familial Case Ascertainment Tool, the best baseline model, were 69 .55%, 0.44%, 65.43%, and 71.12%, respectively) . Our model decreased the number needed to screen compared with the Familial Case Ascertainment Tool (164 versus 227) .
CONCLUSIONS: Our machine learning–derived model provides a higher pretest probability of identifying individuals with a molecular diagnosis of FH compared with current approaches. This provides a promising, cost-effective scalable tool for implementation into electronic health records to prioritize potential FH cases for genetic confirmation.
Key Words: cardiovascular disease prevention ■ familial hypercholesterolemia ■ genetic ■ machine learning ■ screening | | | --- | --- | | Familial hypercholesterolaemia (FH) is a highly prev
alent autosomal codominant genetic condition
occurring with a prevalence of ≈1:311 in the worldwide general population.1,2 FH results in lifelong (from birth) exposure to high low-density lipoprotein cholesterol (LDL-C) levels, increasing the risk of premature | atherosclerotic cardiovascular disease (ASCVD), more notably coronary artery disease (CAD) .1,3 However, if individuals with FH are identified and prescribed effective cholesterol-lowering treatments early, then much of their future adverse health outcomes can be avoided.4 Yet FH remains largely underdiagnosed, with | Correspondence to: Christophe A. T. Stevens, MSc, Department of Primary Care and Public Health, Imperial Centre for Cardiovascular Disease Prevention, Imperial College London, School of Public Health Building, 90 Wood Lane, London W12 7TA, United Kingdom. Email: [christophe.stevens@imperial.ac.uk](christophe.stevens@imperial.ac.uk) *M. T. A. Sharabiani and K. K. Ray are co-senior authors. This manuscript was sent to Manju Bengaluru Jayanna, MD, MS, Assistant Editor, for review by expert referees, editorial decision, and final disposition. Suppleme

FAQs

What problem does the study address in familial hypercholesterolemia detection?

FH is common but largely underdiagnosed, leading to delayed diagnosis and preventable cardiovascular events. The study targets improved identification of potential FH cases for earlier intervention.

How were participants classified as having FH in the analysis?

Participants from UK Biobank were classified as having FH when (likely) pathogenic variants were detected in LDLR, APOB, or PCSK9 genes.

Which machine learning approach performed best and how did it compare with existing criteria?

A stacking ensemble model achieved the best predictive performance and outperformed clinical diagnostic criteria and the UK-recommended screening criteria in identifying FH variant carriers.

Aug 5, 2026

Upload by Ivy