Transformer-Based Multi-Modal Temporal Embeddings for Explainable Metabolic Phenotyping in Type 1 Diabetes

TL;DR

This paper introduces an explainable deep learning framework that combines continuous glucose monitoring and lab data to identify metabolic subgroups in Type 1 Diabetes, revealing distinct biochemical profiles and potential risk factors.

Contribution

It presents a novel multimodal temporal embedding approach using transformers and Gaussian mixture modeling for physiologically meaningful T1D phenotyping.

Findings

Identified five distinct metabolic phenotypes in T1D patients.

Key biomarkers like HbA1c, triglycerides, and TSH differentiate phenotypes.

Glucose variability is a dominant factor influencing metabolic subgroups.

Abstract

Type 1 diabetes (T1D) is a highly metabolically heterogeneous disease that cannot be adequately characterized by conventional biomarkers such as glycated hemoglobin (HbA1c). This study proposes an explainable deep learning framework that integrates continuous glucose monitoring (CGM) data with laboratory profiles to learn multimodal temporal embeddings of individual metabolic status. Temporal dependencies across modalities are modeled using a transformer encoder, while latent metabolic phenotypes are identified via Gaussian mixture modeling. Model interpretability is achieved through transformer attention visualization and SHAP-based feature attribution. Five latent metabolic phenotypes, ranging from metabolic stability to elevated cardiometabolic risk, were identified among 577 individuals with T1D. These phenotypes exhibit distinct biochemical profiles, including differences in glycemic control, lipid metabolism, renal markers, and thyrotropin (TSH) levels. Attention analysis highlights glucose variability as a dominant temporal factor, while SHAP analysis identifies HbA1c, triglycerides, cholesterol, creatinine, and TSH as key contributors to phenotype differentiation. Phenotype membership shows statistically significant, albeit modest, associations with hypertension, myocardial infarction, and heart failure. Overall, this explainable multimodal temporal embedding framework reveals physiologically coherent metabolic subgroups in T1D and supports risk stratification beyond single biomarkers.