CVMar 21, 2025

A-IDE : Agent-Integrated Denoising Experts

arXiv:2503.16780v11 citationsh-index: 1
Originality Incremental advance
AI Analysis

This addresses the challenge of improving CT image quality in heterogeneous, data-scarce medical environments, but it is incremental as it builds on existing RED-CNN models and LLM agents.

The paper tackles the problem of generalizing deep-learning denoising methods across multiple anatomies in Low-Dose CT images by introducing the Agent-Integrated Denoising Experts (A-IDE) framework, which uses an LLM agent to route scans to specialized models, achieving superior performance in RMSE, PSNR, and SSIM on the Mayo-2016 dataset compared to a single model.

Recent advances in deep-learning based denoising methods have improved Low-Dose CT image quality. However, due to distinct HU distributions and diverse anatomical characteristics, a single model often struggles to generalize across multiple anatomies. To address this limitation, we introduce \textbf{Agent-Integrated Denoising Experts (A-IDE)} framework, which integrates three anatomical region-specialized RED-CNN models under the management of decision-making LLM agent. The agent analyzes semantic cues from BiomedCLIP to dynamically route incoming LDCT scans to the most appropriate expert model. We highlight three major advantages of our approach. A-IDE excels in heterogeneous, data-scarce environments. The framework automatically prevents overfitting by distributing tasks among multiple experts. Finally, our LLM-driven agentic pipeline eliminates the need for manual interventions. Experimental evaluations on the Mayo-2016 dataset confirm that A-IDE achieves superior performance in RMSE, PSNR, and SSIM compared to a single unified denoiser.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes