Parameter Efficient Fine-tuning for Domain-specific Gastrointestinal Disease Recognition

2026-04-12

AI Summary

This work addresses the performance degradation in gastrointestinal medical image recognition caused by distribution shifts across different data sources. To tackle this challenge, the authors propose a parameter-efficient fine-tuning approach based on Low-Rank Adaptation (LoRA). By introducing lightweight low-rank matrices into a pre-trained vision foundation model, the method adapts effectively to downstream gastrointestinal disease classification tasks while updating only a small fraction of parameters. Experimental results across multiple datasets demonstrate that LoRA fine-tuning achieves higher classification accuracy than full fine-tuning, despite using significantly fewer trainable parameters, thereby confirming its efficiency and superiority in domain adaptation for medical imaging.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Despite recent advancements in the field of medical image analysis with the use of pretrained foundation models, the issue of distribution shifts between cross-source images largely remains adamant. To circumvent that issue, investigators generally train a separate model for each source. However, this method becomes expensive when we fully fine-tune pretrained large models for a single dataset, as we must store multiple copies of those models. Thus, in this work, we propose using a low-rank adaptation (LoRA) module for fine-tuning downstream classification tasks. LoRAs learn lightweight task-specific low-rank matrices that perturb pretrained weights to optimize those downstream tasks. For gastrointestinal tract diseases, they exhibit significantly better results than end-to-end finetuning with improved parameter efficiency. Code is available at: github.com/sanjay931/peft-gi-recognition.

Problem

• distribution shift
• parameter efficiency
• domain-specific
• gastrointestinal disease recognition
• fine-tuning

Innovation

• Parameter Efficient Fine-tuning
• Low-Rank Adaptation
• Domain-specific Recognition
• Gastrointestinal Disease Classification
• Foundation Models