AutoPrognosis: Automated Clinical Prognostic Modeling via Bayesian Optimization with Structured Kernel Learning

TL;DR

AutoPrognosis is an automated system that designs and optimizes clinical prognostic models using Bayesian optimization and structured kernel learning, improving predictive accuracy and interpretability across diverse healthcare datasets.

Contribution

The paper introduces AUTOPROGNOSIS, a novel automated pipeline for clinical prognosis modeling that combines Bayesian optimization with structured kernel learning and meta-learning for improved performance.

Findings

Outperforms existing prognostic modeling methods across multiple cardiovascular datasets.

Automatically generates interpretable prediction rules for clinicians.

Efficiently optimizes complex model pipelines with high-dimensional hyperparameters.

Abstract

Clinical prognostic models derived from largescale healthcare data can inform critical diagnostic and therapeutic decisions. To enable off-theshelf usage of machine learning (ML) in prognostic research, we developed AUTOPROGNOSIS: a system for automating the design of predictive modeling pipelines tailored for clinical prognosis. AUTOPROGNOSIS optimizes ensembles of pipeline configurations efficiently using a novel batched Bayesian optimization (BO) algorithm that learns a low-dimensional decomposition of the pipelines high-dimensional hyperparameter space in concurrence with the BO procedure. This is achieved by modeling the pipelines performances as a black-box function with a Gaussian process prior, and modeling the similarities between the pipelines baseline algorithms via a sparse additive kernel with a Dirichlet prior. Meta-learning is used to warmstart BO with external data from similar patient cohorts by calibrating the priors using an algorithm that mimics the empirical Bayes method. The system automatically explains its predictions by presenting the clinicians with logical association rules that link patients features to predicted risk strata. We demonstrate the utility of AUTOPROGNOSIS using 10 major patient cohorts representing various aspects of cardiovascular patient care.