Zoish: A Novel Feature Selection Approach Leveraging Shapley Additive Values for Machine Learning Applications in Healthcare.

In the intricate landscape of healthcare analytics, effective feature selection is a prerequisite for generating robust predictive models, especially given the common challenges of sample sizes and potential biases. Zoish uniquely addresses these issues by employing Shapley additive values-an idea rooted in cooperative game theory-to enable both transparent and automated feature selection. Unlike existing tools, Zoish is versatile, designed to seamlessly integrate with an array of machine learning libraries including scikit-learn, XGBoost, CatBoost, and imbalanced-learn.The distinct advantage of Zoish lies in its dual algorithmic approach for calculating Shapley values, allowing it to efficiently manage both large and small datasets. This adaptability renders it exceptionally suitable for a wide spectrum of healthcare-related tasks. The tool also places a strong emphasis on interpretability, providing comprehensive visualizations for analyzed features. Its customizable settings offer users fine-grained control over feature selection, thus optimizing for specific predictive objectives.This manuscript elucidates the mathematical framework underpinning Zoish and how it uniquely combines local and global feature selection into a single, streamlined process. To validate Zoish's efficiency and adaptability, we present case studies in breast cancer prediction and Montreal Cognitive Assessment (MoCA) prediction in Parkinson's disease, along with evaluations on 300 synthetic datasets. These applications underscore Zoish's unparalleled performance in diverse healthcare contexts and against its counterparts.

Meta-Prediction of Coronary Artery Disease Risk.

Coronary artery disease (CAD) remains the leading cause of mortality and morbidity worldwide. Recent advances in large-scale genome-wide association studies have highlighted the potential of genetic risk, captured as polygenic risk scores (PRS), in clinical prevention. However, the current clinical utility of PRS models is limited to identifying high-risk populations based on the top percentiles of genetic susceptibility. While some studies have attempted integrative prediction using genetic and non-genetic factors, many of these studies have been cross-sectional and focused solely on risk stratification. Our primary objective in this study was to integrate unmodifiable (age / genetics) and modifiable (clinical / biometric) risk factors into a prospective prediction framework which also produces actionable and personalized risk estimates for the purpose of CAD prevention in a heterogenous adult population. Thus, we present an integrative, omnigenic, meta-prediction framework that effectively captures CAD risk subgroups, primarily distinguished by degree and nature of genetic risk, with distinct risk reduction profiles predicted from standard clinical interventions. Initial model development considered ~ 2,000 predictive features, including demographic data, lifestyle factors, physical measurements, laboratory tests, medication usage, diagnoses, and genetics. To power our meta-prediction approach, we stratified the UK Biobank into two primary cohorts: 1) a prevalent CAD cohort used to train baseline and prospective predictive models for contributing risk factors and diagnoses, and 2) an incident CAD cohort used to train the final CAD incident risk prediction model. The resultant 10-year incident CAD risk model is composed of 35 derived meta-features from models trained on the prevalent risk cohort, most of which are predicted baseline diagnoses with multiple embedded PRSs. This model achieved an AUC of 0.81 and macro-averaged F1-score of 0.65, outperforming standard clinical scores and prior integrative models. We further demonstrate that individualized risk reduction profiles can be derived from this model, with genetic risk mediating the degree of risk reduction achieved by standard clinical interventions.