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

Ahmed Alaa, Mihaela Schaar
Proceedings of the 35th International Conference on Machine Learning, PMLR 80:139-148, 2018.

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.

Cite this Paper


BibTeX
@InProceedings{pmlr-v80-alaa18b, title = {{A}uto{P}rognosis: Automated Clinical Prognostic Modeling via {B}ayesian Optimization with Structured Kernel Learning}, author = {Alaa, Ahmed and van der Schaar, Mihaela}, booktitle = {Proceedings of the 35th International Conference on Machine Learning}, pages = {139--148}, year = {2018}, editor = {Dy, Jennifer and Krause, Andreas}, volume = {80}, series = {Proceedings of Machine Learning Research}, month = {10--15 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v80/alaa18b/alaa18b.pdf}, url = {http://proceedings.mlr.press/v80/alaa18b.html}, 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.} }
Endnote
%0 Conference Paper %T AutoPrognosis: Automated Clinical Prognostic Modeling via Bayesian Optimization with Structured Kernel Learning %A Ahmed Alaa %A Mihaela Schaar %B Proceedings of the 35th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2018 %E Jennifer Dy %E Andreas Krause %F pmlr-v80-alaa18b %I PMLR %P 139--148 %U http://proceedings.mlr.press/v80/alaa18b.html %V 80 %X 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.
APA
Alaa, A. & Schaar, M.. (2018). AutoPrognosis: Automated Clinical Prognostic Modeling via Bayesian Optimization with Structured Kernel Learning. Proceedings of the 35th International Conference on Machine Learning, in Proceedings of Machine Learning Research 80:139-148 Available from http://proceedings.mlr.press/v80/alaa18b.html.

Related Material