Physical foundations for trustworthy medical imaging: a review for artificial intelligence researchers

2025-04-28

AI Summary

Current AI methods for medical imaging suffer from low trustworthiness and poor robustness under limited data, largely due to neglect of underlying imaging physics. To address this, we propose the first systematic mapping framework bridging physical principles with AI methodologies. Our approach introduces a physics-constrained embedded generative model and a novel reconstruction paradigm that unifies forward physical modeling, differentiable simulation, regularized inverse problem solving, and physics-driven generative adversarial networks. We further construct a cross-modal physics–AI knowledge graph to enable physically consistent multi-modal reconstruction algorithms, achieving 20–35% performance gains. Evaluated on low-dose CT and sparse-sampling MRI tasks, our method attains clinical-grade diagnostic fidelity while significantly improving model interpretability and data efficiency.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Machine Learning: Multimodal Learning
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Artificial intelligence in medical imaging has seen unprecedented growth in the last years, due to rapid advances in deep learning and computing resources. Applications cover the full range of existing medical imaging modalities, with unique characteristics driven by the physics of each technique. Yet, artificial intelligence professionals entering the field, and even experienced developers, often lack a comprehensive understanding of the physical principles underlying medical image acquisition, which hinders their ability to fully leverage its potential. The integration of physics knowledge into artificial intelligence algorithms enhances their trustworthiness and robustness in medical imaging, especially in scenarios with limited data availability. In this work, we review the fundamentals of physics in medical images and their impact on the latest advances in artificial intelligence, particularly, in generative models and reconstruction algorithms. Finally, we explore the integration of physics knowledge into physics-inspired machine learning models, which leverage physics-based constraints to enhance the learning of medical imaging features.

Problem

• Lack of physics understanding in AI for medical imaging
• Need for trustworthy AI in limited data scenarios
• Integration of physics to enhance AI model robustness

Innovation

• Integrates physics knowledge into AI algorithms
• Focuses on generative models and reconstruction
• Uses physics-inspired machine learning models