Application of artificial intelligence for the classification and prediction of pelvic flexion angle after total hip arthroplasty

2021 Fiscal Year Final Research Report

• Project/Area Number: 19K09558
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 56020:Orthopedics-related
• Research Institution: Yokohama City University
• Principal Investigator: INABA Yutaka 横浜市立大学, 医学研究科, 教授 (40336574)
• Co-Investigator(Kenkyū-buntansha): 川上 英良 国立研究開発法人理化学研究所, 科技ハブ産連本部, チームリーダー (30725338)
• Project Period (FY): 2019-04-01 – 2022-03-31
• Keywords: 人工股関節全置換術 / 骨盤傾斜 / 人工知能

Outline of Final Research Achievements

This study aimed to predict the change of pelvic flexion after total hip arthroplasty (THA) using artificial intelligence (AI).
This study involved 415 hips treated with primary THA. Pelvic flexion angle (PFA) was measured, and changes in PFA from preoperative supine position to standing position at 5 years after THA was defined as PFA change (PC). Random Forest (RF) was performed to build the prediction model using explanatory variables including demographic, blood biochemical and radiographic data and the importance of the predictor variables was calculated. Forty-seven hips showed the PC less than -20 degrees that were at the risk of dislocation. Lumbolordotic angle, femoral anteversion angle, BMI, pelvic tilt and sacral slope were most important predictors for PC. Uniform Manifold Approximation and Projection divided these 47 hips into 2 groups, and each group showed different behavior of pelvic tilt.

Free Research Field

関節病学

Academic Significance and Societal Importance of the Research Achievements

本研究より、THA術前の患者因子から術後の骨盤傾斜の予測が高精度で可能となり、THA術前のインプラント設置計画において新たな戦略をもたらすものと考える。
THAは本邦では年々増加し、今後も高齢化社会に伴って増加することを考えると本研究で得られた成果の社会還元は大きく期待され、貢献度も高いものと考える。