Prognostic Stratification of Familial Hypercholesterolemia Patients Using AI Algorithms: A Gender-Specific Approach

AI and Machine Learning (ML) can be highly useful tools for cardiovascular risk stratification in patients with Familial Hypercholesterolemia (FH). Through AI, we can integrate diverse variables such as clinical data, laboratory results, genetic information, imaging findings, and lipid burden, as well as their management through lipid-lowering therapies. In the present study, promising results were obtained, with metrics that outperform the current risk stratification scales used in clinical practice. Using explainability techniques (e.g., SHAP), it was observed that the model variables carry different weights in men and women. Applying explainability models in this new era of AI is crucial to avoid perpetuating gender biases that have historically developed in traditional medicine.