Interpreting deep embeddings for disease progression clustering

TL;DR

This paper introduces a new method for interpreting deep embeddings to cluster patients and gain insights into disease progression, validated on type 2 diabetes data from the UK Biobank.

Contribution

It presents a novel approach for interpreting deep embeddings specifically for patient clustering and disease progression analysis.

Findings

Clinically meaningful insights into disease progression patterns.

Effective clustering of type 2 diabetes patients.

Validation on UK Biobank dataset shows practical utility.

Abstract

We propose a novel approach for interpreting deep embeddings in the context of patient clustering. We evaluate our approach on a dataset of participants with type 2 diabetes from the UK Biobank, and demonstrate clinically meaningful insights into disease progression patterns.