Learning Optimal Summaries of Clinical Time-series with Concept Bottleneck Models
Proceedings of the 7th Machine Learning for Healthcare Conference, PMLR 182:648-672, 2022.
Despite machine learning models’ state-of-the-art performance in numerous clinical prediction and intervention tasks, their complex black-box processes pose a great barrier to their real-world deployment. Clinical experts must be able to understand the reasons behind a model’s recommendation before taking action, as it is crucial to assess for criteria other than accuracy, such as trust, safety, fairness, and robustness. In this work, we enable human inspection of clinical timeseries prediction models by learning concepts, or groupings of features into high-level clinical ideas such as illness severity or kidney function. We also propose an optimization method which then selects the most important features within each concept, learning a collection of sparse prediction models that are sufficiently expressive for examination. On a real-world task of predicting vasopressor onset in ICU units, our algorithm achieves predictive performance comparable to state-of-the-art deep learning models while learning concise groupings conducive for clinical inspection.