Current state of research and future developments of artificial intelligence in pain diagnosis and treatment

Chronic pain represents a significant global health challenge, with traditional diagnostic and therapeutic approaches facing limitations in objective assessment and precision treatment. Advances in artificial intelligence (AI) technology have opened new avenues for addressing these challenges. This review systematically examines the current state and future trends of AI research in the field of pain medicine. The article focuses on three core research directions: In pain assessment, studies concentrate on leveraging AI to fuse multimodal data (such as facial expressions, voice, and physiological signals) to explore its ability to objectively quantify pain states. In image analysis, various deep learning models have been developed to automatically segment key structures like the spine, nerves, and needle tips from medical images, aiming to enhance identification accuracy and efficiency. In disease management, current research explores AI's potential value in classification, treatment decision-making, and prognosis prediction for various painful conditions including shoulder joint disorders, osteoarthritis, trigeminal neuralgia, postherpetic neuralgia, and cancer pain. The article concludes with a critical analysis of challenges in this field-including research data generalization, multimodal fusion strategies, model interpretability, and ethical compliance-while outlining future research directions.