MedIQA: A Scalable Foundation Model for Prompt-Driven Medical Image Quality Assessment

2025-07-25

AI Summary

Existing medical image quality assessment (IQA) methods suffer from poor generalizability, struggling to adapt to multimodal imaging, diverse anatomical regions, and variable-resolution clinical images. To address this, we propose the first foundation model for medical IQA operating across all scales. Our approach innovatively integrates a salient-slice evaluation module with an automatic prompt strategy, enabling alignment between upstream physics-informed pretraining and downstream expert-annotated fine-tuning. Built upon a large-scale multimodal dataset, the model jointly leverages saliency-aware region modeling, prompt-driven learning, and deep neural architectures to deliver cross-modal, cross-anatomical, and resolution-adaptive quality scoring. Extensive experiments demonstrate consistent and significant improvements over state-of-the-art methods across multiple downstream tasks—validating both its strong generalization capability and clinical applicability.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• 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

Rapid advances in medical imaging technology underscore the critical need for precise and automated image quality assessment (IQA) to ensure diagnostic accuracy. Existing medical IQA methods, however, struggle to generalize across diverse modalities and clinical scenarios. In response, we introduce MedIQA, the first comprehensive foundation model for medical IQA, designed to handle variability in image dimensions, modalities, anatomical regions, and types. We developed a large-scale multi-modality dataset with plentiful manually annotated quality scores to support this. Our model integrates a salient slice assessment module to focus on diagnostically relevant regions feature retrieval and employs an automatic prompt strategy that aligns upstream physical parameter pre-training with downstream expert annotation fine-tuning. Extensive experiments demonstrate that MedIQA significantly outperforms baselines in multiple downstream tasks, establishing a scalable framework for medical IQA and advancing diagnostic workflows and clinical decision-making.

Problem

• Automated medical image quality assessment for diagnostic accuracy
• Generalization across diverse modalities and clinical scenarios
• Handling variability in image dimensions, modalities, and regions

Innovation

• First comprehensive foundation model for medical IQA
• Integrates salient slice assessment module for relevant regions
• Employs automatic prompt strategy aligning pre-training with fine-tuning