Artificial intelligence in microsurgery and supermicrosurgery training within plastic surgery: A systematic review.

Introduction: Artificial Intelligence (AI) is increasingly integrated into plastic surgery. Technical complexities of microsurgery and supermicrosurgery demand incorporation of AI in facilitating steep learning curves. Although simulation models exist, many provide inadequate preparation and fidelity. AI has potential to complement trainee efforts in skill acquisition by improving training efficacy. This systematic review aims to present and evaluate applications and future of AI in microsurgery and supermicrosurgery plastic surgery training. Materials and Methods: PubMed, Embase, Scopus and Google Scholar were searched following PRISMA guidelines. Studies on AI applications on microsurgery and supermicrosurgery training within plastic surgery were included. Two independent reviewers screened the literature, extracted data and assessed risk of bias using ROBINS-I tool. AI model architecture and motion, eye and instrument tracking parameters and narrative data synthesis were captured on Microsoft Excel. (PROSPERO ID: CRD42025607695). Results: Five articles were included. Two articles developed automated microsurgical skill classification datasets with force-based surgical glove model achieving above 95 % accuracy. Two AI applications used eye tracking parameters namely, blink rate, pupil dilation and gaze coordination to assess proficiency and workload. Path length, mean velocity and jerk curvature were accurately analyzed by AI models with instrument and hand tracking. All studies reported moderate risk of bias. Conclusion: This is the first systematic review addressing AI applications in microsurgery and supermicrosurgery training. AI models tracking eye, hand and instrument motion for microsurgical skill assessment demonstrated accuracy and precision, enabling real-time monitoring of proficiency and workload, which are scalable, objective methods for improving training and outcomes.