| I&T Solution |
AI-Driven Rehabilitation Training and Assessment System with Multi-Sensor Fused Wearable Devices
(REF: S-1447) |
| Trial Project |
|
| Solution Feature |
- Multiple wearable sensors. These sensors capture real-time data during rehabilitation exercises of the elderly, which is then used to generate analysis reports.
- AI-based data analysis. We can analyze the data and obtain the body posture and limp detection results and provide a quality score for the completion of training.
- Rehabilitation training recommendations. Combing collected data with the clinical experience the therapists, rehabilitation recommendations are provided.
- Remote data access. Therapists can access the cloud-based analysis results and establish network with the elderly for real-time clinical assessment.
- User-friendly mobile application interface. The software design greatly reducs the learning curve for the elderly to use the software.
|
| Trial Application and Expected Outcome |
- Features of the mobile application include: 1) real-time data visualization; 2) real-time communication functionality for remote data access by therapists.
- Body pose estimation and visualization. Reconstruct high-resolution 3D meshes of the body posture. Expected accuracy is up to 98% at a threshold of 15mm.
- Limp detection. An AI algorithm is used to determine whether the elderly person has a limp. Expected detection accuracy is above 95%.
- Walking speed detection. The system calculates and displays the elderly person's walking speed in real-time using intelligent algorithms.
- Rehabilitation training performance scoring. The application assesses the training quality. It displays a dashboard score indicating the quality.
|
| Info on I&T Solution Provider |
| Solution Provider | : | GONG Leiliang | | Address | : | HKPC Building, 78 Tat Chee Avenue, Kowloon, Hong Kong. | | Contact Person | : | GONG Leiliang |
| Position | : | Consultant | | Tel | : | 27885653 | | Email | : |
leiliang@hkpc.org | | Webpage | : | https://www.hkpc.org/zh-HK |
|
