(0507) Artificial Intelligence for Outcome Prediction and Evaluation in Facial Plastic Surgery: A Systematic Review

Introduction: The utilization of artificial intelligence (AI) in facial plastic surgery, by both providers and patients, has increased in prevalence within the last few years. In this systematic review, we seek to assess the current utilization of AI for outcome prediction and evaluation in the field of cosmetic facial plastic surgery.

Methods: The review was conducted in adherence to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines. Medline, PubMed, Scopus, and Cochrane databases were searched from January 1982 to October 2023 using keywords relevant to cosmetic surgeries and artificial intelligence. Inclusion and exclusion criteria were used to compile eligible articles. Quality and risk of bias of selected articles were evaluated using the QUADAS-2 tool for diagnostic accuracy and NIH tool for before and after studies.

Results: In total, 1,274 publications were identified from databases. Of these, 14 publications satisfied inclusion criteria. Publications evaluated the usage of AI or AI photo enhancers in outcome prediction (n = 3) or evaluation (n = 11) in rhinoplasty, blepharoplasty, and facelift. Two studies assessed using the QUADAS-2 tool had low risk of bias, and two studies assessed using the NIH tool demonstrated high quality. Multiple studies utilized reduction in AI-perceived age (n = 5) or facial expression of emotion (n = 3) as measures of outcome evaluation and postoperative simulation (n = 2) as methods of outcome prediction. Included studies found reduction in AI-perceived age and change in expression of emotion following cosmetic surgery. High accuracy of AI-estimated age and postoperative outcome simulation was demonstrated.

Conclusions: While this review provides promising insight into the applications, benefits, and accuracy of AI and AI photo enhancers, current research is limited and a conclusion for its integration into care cannot be drawn due to the limited quantity of existing studies. The usage of AI in facial plastic surgery has the potential for exponential growth in the coming years, which warrants further evaluation and ethical considerations by surgeons.