AI Posture Detection With Multi-Camera 3D Joint Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current physiotherapy systems lack real-time quantification and comprehensive assessment capabilities, particularly in home settings, failing to provide accurate injury detection, therapeutic guidance, and frailty evaluation, and are limited by two-dimensional pose detection.
Innovation Solution
A system combining computer vision algorithms, AI models, and statistical tools to convert high-dimensional healthcare monitoring videos into low-dimensional data for real-time physiotherapy, fall detection, and frailty assessment, using three-dimensional joint identification and digital twin technology for personalized therapy guidance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If two-dimensional pose detection is used, then device complexity is reduced, but measurement precision and functionality are severely limited
Solution Approach 1:
The patent transitions from two-dimensional pose detection to three-dimensional pose detection by incorporating depth information through multiple cameras arranged in specific geometries. This dimensional enhancement allows the system to accurately capture spatial relationships, joint angles, and body orientation in real-time, significantly improving measurement precision while maintaining manageable system complexity through standardized camera configurations and processing algorithms.
2Productivity
If real-time video analysis is performed, then productivity increases, but device complexity and computational requirements worsen
Solution Approach 1:
The patent divides the complex video analysis task into separate functional modules: pose detection module, metric calculation module, and assessment module. Each module processes specific aspects of the video data independently, then integrates results to provide comprehensive real-time assessments. This segmentation enables parallel processing and reduces computational burden on any single component while maintaining high productivity.
Solution Approach 2:
The system creates a virtual 3D model (digital twin) of the patient's body from video data, which serves as a simplified representation for analysis. This digital copy can be processed repeatedly without re-analyzing the original video, reducing computational complexity while maintaining assessment accuracy and enabling real-time feedback.
3Measurement precision
If comprehensive assessment metrics are implemented, then measurement precision improves, but device complexity and data processing requirements worsen
Solution Approach 1:
The patent implements a universal assessment framework that can evaluate multiple parameters (joint angles, range of motion, gait speed, balance, frailty indicators) using the same 3D pose detection infrastructure. This multi-functional approach allows comprehensive assessments without proportionally increasing system complexity, as the core technology serves all assessment purposes through configurable analysis parameters.
4Measurement precision
If quantitative metrics are collected, then measurement precision improves, but loss of information increases due to high-dimensional data
Solution Approach 1:
The patent extracts and isolates specific quantitative metrics of interest from the high-dimensional video data, such as joint angles, range of motion, gait parameters, and balance measures. By focusing extraction on clinically relevant parameters rather than processing all raw pixel data, the system maintains measurement precision for critical assessments while reducing the dimensional complexity of stored and processed information.
Data Source
AI summary
An AI-assisted, data driven, real time posture detection system for physiotherapy, fall prevention, and/or frailty assessments. The system uses a pre-trained pose detection model and video images from multiple angles captured by multiple cameras to determine the three-dimensional locations of user joints in real time. In embodiments, the system calculates qualitative metrics (e.g., range of motion, ankle dorsiflexion, Q-angle, hip-knee-ankle alignment, gait speed, etc.) to determine whether the user has suffered or is at risk of an injury (e.g., ACL tear, patellar tendonitis, hip fracture, etc.) and/or whether a physiological condition has worsened or improved over time (e.g., after surgery or physical therapy). In some embodiments, the system constructs a digital twin of the user and displays a visual representation of the digital twin performing idealized movements (e.g., proper form for exercise or physical therapy) to provide real-time instruction and feedback to improve the physiological condition of the user.