Abstract:

Objective　To explore the clinical effect of AI visualization technology in preoperative planning of complex total hip arthroplasty. Methods　A retrospective analysis was performed on 62 patients who underwent complex total hip arthroplasty in Huangshi Central Hospital from July 2020 to April 2023. The patients were divided into two groups according to different THA preoperative planning methods: 25 patients undergoing preoperative planning using AI visualization technology were included in the experimental group [12 males and 13 females, aged (67.00±11.34) years], and 37 patients undergoing preoperative planning using traditional two-dimensional X-ray film were included in the control group [18 males and 19 females, aged (66.89±12.98) years]. The operation time, intraoperative blood loss, intraoperative fluoroscopy times, first postoperative exercise time, postoperative length difference between the two lower limbs, Harris Hip Score (HHS) and Visual Analogue Scale (VAS) score before surgery, 3 months after surgery, and 1 year after surgery, and the coincidence rates between the size of the acetabular and femur prostheses actually used during surgery and the size of the preoperative planned prostheses were recorded in both groups. t test and χ² test were used. Results　The operation time, intraoperative blood loss, intraoperative X-ray fluoroscopy times, and postoperative length difference between the two lower limbs of the experimental group were all better than those of the control group [(81.08±6.31) min vs. (102.35±8.14) min, (408.44±29.83) ml vs. (465.58±72.09) ml, (1.44±0.51) times vs. (2.24±0.64) times, (3.28±0.46) mm vs. (4.24±0.72) mm], with statistically significant differences (t=11.01, 3.75, 5.25, and 5.90, all P<0.05). The HHS scores at 3 months and 1 year after surgery of the experimental group were higher than those of the control group, the VAS scores at 3 months and 1 year after surgery were lower than those of the control group, and the first postoperative exercise time was shorter than that of the control group, with statistically significant differences (t=-7.38, -7.13, 4.24, 4.79, and 3.58, all P<0.05). The coincidence rates between the size of the acetabular and femur prostheses actually used during surgery and the size of the preoperative planned prostheses in the experimental group were higher than those in the control group [88.00% (22/25) vs. 56.76% (21/37), 92.00% (23/25) vs. 54.05% (20/37)], with statistically significant differences (χ²=6.85 and 10.11, both P<0.05). Conclusion　Surgical planning according to AI visualization technology before complex total hip arthroplasty can improve the prostheses size coincidence rate of hip joint on acetabular side and femur side, reduce the operation time, intraoperative blood loss, and fluoroscopy times, improve the hip joint function, and relieve the pain.

Key words:

Total hip arthroplasty, Artificial intelligence, Visualization, Preoperative planning

摘要：

目的　探索AI可视化技术在复杂全髋关节置换术（THA）患者术前规划中的临床效果。方法　选取2020年7月至2023年4月在黄石市中心医院行复杂THA的62例患者进行回顾性分析。根据THA不同术前规划方法分为两组，采用AI可视化技术进行术前规划的25例患者为试验组［男12例、女13例，年龄（67.00±11.34）岁］，应用传统二维X线胶片行术前规划的37例患者为对照组［男18例、女19例，年龄（66.89±12.98）岁］。记录两组患者手术时间、术中出血量、术中X线透视次数、术后首次下地活动时间、术后双下肢长度差值，术前、术后3个月及术后1年Harris髋关节评分系统（HHS）评分、视觉模拟评分法（VAS）评分，术中实际髋臼及股骨侧使用假体大小型号与术前规划假体型号的符合率。统计学方法采用t检验、χ²检验。结果　试验组手术时间、术中出血量、术中X线透视次数、术后双下肢长度差值均优于对照组［（81.08±6.31）min比（102.35±8.14）min、（408.44±29.83）ml比（465.58±72.09）ml、（1.44±0.51）次比（2.24±0.64）次、（3.28±0.46）mm比（4.24±0.72）mm］，差异均有统计学意义（t=11.01、3.75、5.25、5.90，均P<0.05）。试验组术后3个月、1年HHS评分均高于对照组，术后3个月、1年VAS评分均低于对照组，术后首次下地时间短于对照组，差异均有统计学意义（t=-7.38、-7.13、4.24、4.79、3.58，均P<0.05）。试验组术中实际使用髋臼侧及股骨侧假体大小型号符合率均高于对照组［88.00%（22/25）比56.76%（21/37）、92.00%（23/25）比54.05%（20/37）］，差异均有统计学意义（χ²=6.85、10.11，均P<0.05）。结论　复杂THA术前按照AI可视化技术进行手术规划可提高髋关节髋臼侧及股骨侧假体型号吻合率，减少手术时间、术中出血量及透视次数，改善患者术后髋关节功能，缓解疼痛。

关键词:

全髋关节置换术, 人工智能, 可视化, 术前规划