TopoAgent: An Agentic Framework for Automated Topology Learning in Medical Imaging

2026-06-29

AI Summary

This work addresses the limitations of existing methods that rely on fixed topological descriptors and struggle to adapt to the diverse topological structures present in medical images. The authors propose the first large language model (LLM) agent framework that operates without task-specific training, leveraging a perception–reasoning–action–reflection loop to automatically select optimal topological descriptors and generate informative feature vectors. The framework integrates persistent homology, fifteen distinct topological descriptors, twenty-one domain-specific tools, and a dual-memory mechanism to enable cross-dataset experience transfer. Extensive experiments across twenty-six medical image datasets demonstrate that the proposed approach significantly enhances the adaptability of topological feature extraction and improves downstream task performance.

Abstract

Topological data analysis (TDA), particularly persistent homology (PH), captures geometric structural properties in medical images (e.g., connected components, loops, shape characteristics), which conventional pixel-level deep learning approaches often neglect. While many topological descriptors are known for converting persistence diagrams (PDs) or raw images into topological feature vectors, existing methods mostly default to a single fixed descriptor (e.g., persistence images), leaving the diversity of topological representations largely unexplored. To the best of our knowledge, there is no known large language model (LLM)-based agentic framework that can automatically determine the most suitable topological descriptors for a given image dataset and produce the corresponding topological feature vectors for downstream tasks. To fill this gap, we propose \textbf{TopoAgent}, an LLM-based agentic framework that automates topology learning for medical image analysis.TopoAgent operates through a Perception--Reasoning--Action--Reflection loop supported by 21 domain-specific tools and dual memory that accumulates experience across runs. Its skill set is distilled from systematic evaluation of 15 topological descriptors across 26 datasets with six classifiers. TopoAgent analyzes input images and their topological characteristics, reasons about which topological descriptors best suit the input, and determines the optimal descriptor and its configuration, all without task-specific training.

Problem

• topological data analysis
• persistent homology
• medical imaging
• topological descriptors
• automated topology learning

Innovation

• Topological Data Analysis
• Large Language Model
• Automated Topology Learning
• Persistent Homology
• Medical Image Analysis