Incorporating Metabolic Information into LLMs for Anomaly Detection in Clinical Time-Series

TL;DR

The Metabolism Pathway-driven Prompting method is introduced, which integrates the information about metabolic pathways to better capture the structural and temporal changes in biological samples to improve anomaly detection performance.

Abstract

Anomaly detection in clinical time-series holds significant potential in identifying suspicious patterns in different biological parameters. In this paper, we propose a targeted method that incorporates the clinical domain knowledge into LLMs to improve their ability to detect anomalies. We introduce the Metabolism Pathway-driven Prompting (MPP) method, which integrates the information about metabolic pathways to better capture the structural and temporal changes in biological samples. We applied our method for doping detection in sports, focusing on steroid metabolism, and evaluated using real-world data from athletes. The results show that our method improves anomaly detection performance by leveraging metabolic context, providing a more nuanced and accurate prediction of suspicious samples in athletes' profiles.

Figures (9)

• Figure 1: Simplified steroid pathway.
• Figure 2: Schematic diagram of Metabolism Pathway-driven Prompting (MPP) method.
• Figure 3: t-SNE representation of embeddings of the output from different prompting methods.
• Figure 4: Example Prompt for Zero-shot Learning.
• Figure 5: Example Prompt for In-context Learning.