ProMedTS: A Self-Supervised, Prompt-Guided Multimodal Approach for Integrating Medical Text and Time Series

TL;DR

ProMedTS introduces a self-supervised, prompt-guided multimodal framework that effectively integrates medical time series data with clinical notes, improving disease diagnosis accuracy by aligning heterogeneous data in a shared representation space.

Contribution

This work presents a novel self-supervised approach using prompt-guided learning and anomaly detection to unify time series and text data in medical applications, which has not been explored before.

Findings

ProMedTS outperforms existing methods on disease diagnosis tasks.

The framework effectively aligns temporal and semantic information.

Prompt-guided learning enhances multimodal data integration.

Abstract

Large language models (LLMs) have shown remarkable performance in vision-language tasks, but their application in the medical field remains underexplored, particularly for integrating structured time series data with unstructured clinical notes. In clinical practice, dynamic time series data, such as lab test results, capture critical temporal patterns, while clinical notes provide rich semantic context. Merging these modalities is challenging due to the inherent differences between continuous signals and discrete text. To bridge this gap, we introduce ProMedTS, a novel self-supervised multimodal framework that employs prompt-guided learning to unify these heterogeneous data types. Our approach leverages lightweight anomaly detection to generate anomaly captions that serve as prompts, guiding the encoding of raw time series data into informative prompt embeddings. These prompt embeddings are aligned with textual representations in a shared latent space, preserving fine-grained temporal nuances alongside semantic insights. Furthermore, our framework incorporates tailored self-supervised objectives to enhance both intra- and inter-modal alignment. We evaluate ProMedTS on disease diagnosis tasks using real-world datasets, and the results demonstrate that our method consistently outperforms state-of-the-art approaches.