SWoTTeD: An Extension of Tensor Decomposition to Temporal Phenotyping
This addresses the problem of interpreting temporal data in healthcare for clinicians, representing an incremental improvement with novel constraints for interpretability.
The paper tackles the challenge of analyzing complex temporal patterns in individual traces like Electronic Health Records by introducing SWoTTeD, a method for temporal tensor decomposition that discovers hidden temporal phenotypes; results show it achieves at least as accurate reconstruction as state-of-the-art models and extracts clinically meaningful phenotypes.
Tensor decomposition has recently been gaining attention in the machine learning community for the analysis of individual traces, such as Electronic Health Records (EHR). However, this task becomes significantly more difficult when the data follows complex temporal patterns. This paper introduces the notion of a temporal phenotype as an arrangement of features over time and it proposes SWoTTeD (Sliding Window for Temporal Tensor Decomposition), a novel method to discover hidden temporal patterns. SWoTTeD integrates several constraints and regularizations to enhance the interpretability of the extracted phenotypes. We validate our proposal using both synthetic and real-world datasets, and we present an original usecase using data from the Greater Paris University Hospital. The results show that SWoTTeD achieves at least as accurate reconstruction as recent state-of-the-art tensor decomposition models, and extracts temporal phenotypes that are meaningful for clinicians.