Explaining a machine learning decision to physicians via counterfactuals

TL;DR

This paper introduces a VAE-based method for generating plausible, relevant, and sparse counterfactual explanations for time-series data to improve interpretability of machine learning models in healthcare.

Contribution

It proposes a novel VAE approach that produces realistic and meaningful counterfactuals for clinical time-series data, addressing prior limitations in plausibility and speed.

Findings

Generated counterfactuals are more plausible and relevant according to physicians.

The method is 100 times faster than previous approaches.

Quantitative evaluation shows improved likelihood of counterfactuals within data distribution.

Abstract

Machine learning models perform well on several healthcare tasks and can help reduce the burden on the healthcare system. However, the lack of explainability is a major roadblock to their adoption in hospitals. \textit{How can the decision of an ML model be explained to a physician?} The explanations considered in this paper are counterfactuals (CFs), hypothetical scenarios that would have resulted in the opposite outcome. Specifically, time-series CFs are investigated, inspired by the way physicians converse and reason out decisions `I would have given the patient a vasopressor if their blood pressure was lower and falling'. Key properties of CFs that are particularly meaningful in clinical settings are outlined: physiological plausibility, relevance to the task and sparse perturbations. Past work on CF generation does not satisfy these properties, specifically plausibility in that realistic time-series CFs are not generated. A variational autoencoder (VAE)-based approach is proposed that captures these desired properties. The method produces CFs that improve on prior approaches quantitatively (more plausible CFs as evaluated by their likelihood w.r.t original data distribution, and 100$\times$ faster at generating CFs) and qualitatively (2$\times$ more plausible and relevant) as evaluated by three physicians.