Rafael G. JAKUBIETZ, Rainer H. MEFFERT, Tabea FIEBER, Michael G. JAKUBIETZ
Turkish Journal of Plastic Surgery - 2026;34(2):68-74
Recent advancements in artificial intelligence (AI) have a major influence on some aspects of health care. AI, with its 4 pillars of natural language processing (NLP), machine learning (ML), computer vision, and robotics, has advanced significantly in the past decades. NLP with large language models (LLMs) has created widespread awareness in the general public in recent years and could be useful in many fields besides medicine. The most important application of AI in current medical practice is found in the subcategory of computer vision, with a focus on image-related specialties. The field of radiology has been at the forefront of clinical application, with 75% of Food and Drug Administration-cleared AI algorithms aimed at radiology practice. Contrary to this, much clinical data relevant to surgical fields, such as plastic and hand surgery, consists of mostly subjective parameters and lacks standardization. Despite this, AI offers the potential to improve diagnostics, support surgical decision-making, and predict benefits or complications through pattern recognition. This retrospective literature review summarizes current applications relevant to the clinical field of hand surgery. AI-driven imaging analysis may allow rapid detection of complex fracture-dislocations facilitating early and accurate diagnosis. Large language models have also been successfully used to create postoperative rehabilitation protocols providing improved access to hand rehabilitation. The literature concerning surgical planning and decision-making, robotic and bionic applications specific to hand surgery is less extensive. Machine learning algorithms assisting in medical documentation are currently more advanced in clinical implementation. The integration of artificial intelligence into hand surgery promises improved patient care in the future, but many challenges remain regarding ethics, data security, and reliability.
