Clusters in Explanation Space: Inferring disease subtypes from model explanations
This work addresses the challenge of identifying disease subtypes for improved clinical diagnosis and treatment, though it appears incremental as it adapts existing explanation and clustering techniques to a specific domain.
The authors tackled the problem of discovering disease subtypes from noisy biomedical data by proposing a new approach that clusters instances in the explanation space of a diagnostic classifier, rather than the original data. They showed that this method substantially outperforms classical cluster analysis on original data in recovering known subtypes, as demonstrated on UK Biobank brain imaging and Cancer Genome Atlas transcriptome datasets.
Identification of disease subtypes and corresponding biomarkers can substantially improve clinical diagnosis and treatment selection. Discovering these subtypes in noisy, high dimensional biomedical data is often impossible for humans and challenging for machines. We introduce a new approach to facilitate the discovery of disease subtypes: Instead of analyzing the original data, we train a diagnostic classifier (healthy vs. diseased) and extract instance-wise explanations for the classifier's decisions. The distribution of instances in the explanation space of our diagnostic classifier amplifies the different reasons for belonging to the same class - resulting in a representation that is uniquely useful for discovering latent subtypes. We compare our ability to recover subtypes via cluster analysis on model explanations to classical cluster analysis on the original data. In multiple datasets with known ground-truth subclasses, most compellingly on UK Biobank brain imaging data and transcriptome data from the Cancer Genome Atlas, we show that cluster analysis on model explanations substantially outperforms the classical approach. While we believe clustering in explanation space to be particularly valuable for inferring disease subtypes, the method is more general and applicable to any kind of sub-type identification.