A study on the feasibility and initial outcomes of artificial intelligence-assisted preoperative planning for hip revision surgery

Objective The aim of this study was to explore the early and midterm efficacy of artificial intelligence (AI) involvement in assisted hip revision surgery and to summarize our clinical experience. Methods Clinical data were collected from 38 patients (39 hips) who underwent hip revision surgery and preoperative planning using AI technology in our hospital between June 2019 and November 2024. The cohort included 17 men and 22 women, with a mean age of 70.33 ± 10.42 years (range 44–90 years). Mean follow-up time was 39.67 ± 16.76 months (range 7–72 months). The initial revision was performed in 37 hips, second revision in three hips, and third revision in one hip. The function of the affected limb was evaluated using Harris Hip Scoring System (HHS) at preoperative and postoperative nodes. Results In this study, 38 patients were monitored for 7–72 months, with a mean follow-up of 34.64 ± 16.54 months. Matching of AI-planned prosthesis size with actual prosthesis size replaced during surgery: 3 cases (7.69%) showed a ±1 size discrepancy, 1 case (2.56%) showed a ±2 size discrepancy, and the remaining patients achieved precise matching. The HHS scores of all patients at 6 months postoperatively (82.99 ± 6.91) were higher than the preoperative scores (33.03 ± 7.36), and the difference was statistically significant ( P < 0.05). Conclusion AI technology makes complex revision surgeries simple by accurately formulating individual preoperative plans for hip revision surgery. The feasibility and preliminary efficacy of AI-assisted revision total hip arthroplasty surgery are satisfactory. AI-assisted complex orthopedic surgeries warrant further in-depth clinical research.