Related Experiment Video
Updated: Jun 6, 2025
08:08
Cell-based Assay Protocol for the Prognostic Prediction of Idiopathic Scoliosis Using Cellular Dielectric Spectroscopy
Published on: October 16, 2013
10.6K
Intelligent Evaluation Method for Scoliosis at Home Using Back Photos Captured by Mobile Phones.
Yongsheng Li1, Xiangwei Peng2, Qingyou Mao3
1Institute for Hospital Management of Tsinghua University, Shenzhen 518000, China.
Bioengineering (Basel, Switzerland)
|November 27, 2024
Summary
This study introduces a novel computer vision method using mobile phone photos for scoliosis screening and rehabilitation monitoring. The approach accurately assesses spinal curvature, deviation, and trunk rotation, outperforming other deep learning techniques.
Area of Science:
- Biomedical Engineering
- Computer Vision
- Orthopedics
Background:
- Traditional X-ray screening for scoliosis is impractical for large-scale use and dynamic rehabilitation monitoring.
- A non-invasive, accessible method for scoliosis assessment is needed.
Purpose of the Study:
- To develop and validate a computer vision-based method for evaluating spinal curvature using mobile phone back images.
- To quantify spinal deviation in both coronal and sagittal planes and measure trunk rotation.
Main Methods:
- Utilized YOLOv8 key point detection for spinal coronal curvature classification.
- Developed an algorithm to quantify spinal coronal plane deviation using back key points.
- Implemented a multi-scale automatic peak detection (AMPD) algorithm to measure trunk rotation (ATR angle) for sagittal plane deviation.
Main Results:
- The proposed method demonstrated high accuracy in classifying scoliosis types and quantifying spinal deviation.
- Performance evaluation using public and clinical datasets showed superior effectiveness compared to existing deep learning algorithms.
- The system successfully correlated mobile phone image analysis with X-ray data.
Conclusions:
- This computer vision approach offers an accurate and effective alternative for scoliosis assessment.
- The method is suitable for both initial screening and dynamic monitoring during rehabilitation.
- Mobile phone-based analysis presents a promising, accessible tool for scoliosis management.
