Clinical efficacy of artificial intelligence-based pelvic floor optimization training system using human body motion analysis technique
10.12483/j.issn.1009-8291.2026.06.005
- VernacularTitle:基于人体运动分析技术的人工智能盆底优化训练系统的临床应用疗效研究
- Author:
Xiaochen LUO
1
;
Jialing YAO
2
;
Jinle XIE
2
;
Qiang WU
1
;
Jiayang HE
3
;
Yangyun WANG
2
Author Information
1. Department of Equipment, The Fifth People's Hospital of Shanghai, Fudan University, Shanghai 200240, China
2. Pelvic Floor Center, The Fifth People's Hospital of Shanghai, Fudan University, Shanghai 200240, China
3. Department of Urology, The Fifth People's Hospital of Shanghai, Fudan University, Shanghai 200240, China
- Publication Type:Journal Article
- Keywords:
overactive bladder;
AI-based pelvic floor optimization training system;
pelvic floor rehabilitation;
human motion analysis
- From:
Journal of Modern Urology
2026;31(6):528-535
- CountryChina
- Language:Chinese
-
Abstract:
Objective To apply an artificial intelligence (AI) -based pelvic floor optimization training system using human motion analysis technique in patients undergoing pelvic floor rehabilitation, so as to assist clinicians in providing guidance, improving clinical efficiency, and enhancing the efficacy of patients' independent training. Methods A total of 60 patients with overactive bladder (OAB) complicated with urgent urinary incontinence (UUI) treated in our hospital during Sep. 2024 and Dec. 2025 were enrolled and randomly divided into the routine guidance group (n=30) and AI guidance group (n=30). The routine guidance group received conventional pelvic floor optimization training under clinicians' instruction, while the AI guidance group underwent training with the AI-based pelvic floor optimization system using human motion analysis. After 6 weeks of intervention, outcomes were compared between the two groups, including the international consultation on incontinence questionnaire-short form (ICI-Q-SF), incontinence quality of life scale (I-QoL), overactive bladder symptom score (OABSS), patient's perception of bladder condition (PPBC) score, pelvic floor muscle function (myoelectric values in the preresting, post-resting, type Ⅰ muscle, type Ⅱ muscle, and endurance test phases), urine output, average flow rate (Qavg), and maximum flow rate (Qmax). Results After 6 weeks of treatment, the ICI-Q-SF, I-QoL, OABSS, PPBC score, type I muscle myoelectric value, Qavg and Qmax were significantly improved in both groups compared with baseline (P＜0.05). After 6 weeks of intervention, the I-QoL score was significantly higher in the AI guidance group than in the routine guidance group[81.82 (79.55, 84.38) vs.77.84 (75.00, 79.55) ], with significant difference (P＜0.05). Compared with the routine guidance group, the AI guidance group showed significant improvements in myoelectric values in the pre-resting, post-resting, type I muscle and endurance testing phases (P＜0.05). Conclusion The AI-based pelvic floor optimization training system can effectively assist clinicians in guiding patients to complete pelvic floor optimization training, improve rehabilitation efficiency and efficacy of independent exercise, and reduce dependence on specialized clinicians, which exhibits favorable safety.
