Cross-race vs. same-race collaboration in plastic surgery: AI-driven analysis of research and race-interaction dynamics

Background: Racial diversity in plastic surgery research remains a challenge. This study employs artificial intelligence (AI)-driven big data analysis to assess racial representation, cross-race mentorship trends, and research productivity disparities among plastic surgery authors over 12 years (2010–2022). Methods: A retrospective, cross-sectional study was conducted using AI-based web scraping to extract authorship data from 24 PubMed-indexed plastic surgery journals. RaceAPI, a machine-learning-based name-classification tool trained on U.S. Census data, inferred racial identity. Chi-square tests were used to evaluate racial disparities in first and senior authorship, mentorship patterns, and research output. Results: Among 16 019 publications, White researchers remained the dominant group among first authors (70% in 2010 vs. 66.2% in 2022, P < 0.001) and senior authors (73.8% vs. 72.6%, P < 0.01). Black first and senior authorship declined (8.1% to 6.4%, P < 0.001; 9.2% to 6.1%, P < 0.001), while Asian representation increased (first: 16% to 18.8%, P = 0.027; senior: 11.7% to 16%, P < 0.001). White authors had significantly higher research productivity (2.15 papers per author) than Asian (1.97), Black (1.76), and Hispanic (1.68) researchers ( P < 0.05). Black senior authors had the highest rate of cross-race mentorship (77.5%), while White (26.8%) and Asian (27.3%) senior authors exhibited strong same-race mentorship preferences. Conclusions: Persistent racial disparities in authorship and research participation among racial groups that are persistently underrepresented in medical school and plastic surgery residencies suggest the need to assess and address educational barriers prior to medical school. Strengthening mentorship networks and increasing access to research for minority trainees could foster a thriving plastic surgery workforce for all communities, including those currently underserved. Level of Evidence: Level III, cross-sectional study.