CTG-Insight: A Multi-Agent Interpretable LLM Framework for Cardiotocography Analysis and Classification

2025-07-29

AI Summary

In remote fetal monitoring, the lack of interpretability in cardiotocography (CTG) data hinders maternal comprehension and clinical trust. To address this, we propose a multi-agent large language model (LLM)-based framework for interpretable CTG analysis. Grounded in clinical guidelines, CTG signals are decomposed into five physiological dimensions—baseline heart rate, variability, accelerations, decelerations, and sinusoidal patterns—each analyzed in parallel by dedicated domain-specific agents. A central aggregation agent then synthesizes findings into structured diagnostic conclusions and natural-language explanations. Our approach achieves both high accuracy and strong interpretability: 96.4% accuracy and 97.8% F1-score on the NeuroFetalNet dataset—significantly outperforming single-agent baselines and state-of-the-art deep learning models. To our knowledge, this is the first fully automated, modular, guideline-driven framework enabling interpretable, end-to-end CTG interpretation.

Technology Category

• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Multiagent Systems: Modeling other Agents

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Remote fetal monitoring technologies are becoming increasingly common. Yet, most current systems offer limited interpretability, leaving expectant parents with raw cardiotocography (CTG) data that is difficult to understand. In this work, we present CTG-Insight, a multi-agent LLM system that provides structured interpretations of fetal heart rate (FHR) and uterine contraction (UC) signals. Drawing from established medical guidelines, CTG-Insight decomposes each CTG trace into five medically defined features: baseline, variability, accelerations, decelerations, and sinusoidal pattern, each analyzed by a dedicated agent. A final aggregation agent synthesizes the outputs to deliver a holistic classification of fetal health, accompanied by a natural language explanation. We evaluate CTG-Insight on the NeuroFetalNet Dataset and compare it against deep learning models and the single-agent LLM baseline. Results show that CTG-Insight achieves state-of-the-art accuracy (96.4%) and F1-score (97.8%) while producing transparent and interpretable outputs. This work contributes an interpretable and extensible CTG analysis framework.

Problem

• Improves interpretability of fetal monitoring CTG data
• Classifies fetal health using multi-agent LLM analysis
• Achieves high accuracy with transparent medical explanations

Innovation

• Multi-agent LLM system for CTG analysis
• Decomposes CTG into five medical features
• Achieves high accuracy with interpretable outputs