Related Experiment Video
Updated: Jul 2, 2025
08:08
Cell-based Assay Protocol for the Prognostic Prediction of Idiopathic Scoliosis Using Cellular Dielectric Spectroscopy
Published on: October 16, 2013
10.7K
Deep learning algorithm for automatically measuring Cobb angle in patients with idiopathic scoliosis
Ming Xing Wang1, Jeoung Kun Kim1, Jin-Woo Choi2
1Department of Business Administration, School of Business, Yeungnam University, Gyeongsan-Si, Republic of Korea.
Summary
A new deep learning model automatically measures the Cobb angle for scoliosis, significantly reducing measurement variability and errors. This AI tool offers excellent reliability for clinical practice.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Orthopedics
Background:
- The Cobb angle is crucial for scoliosis assessment but suffers from high inter- and intra-observer variability.
- Measurement inconsistencies can arise from variations in vertebral identification and measurement techniques.
Purpose of the Study:
- To develop and validate a deep learning model for automated Cobb angle measurement.
- To improve the accuracy and consistency of scoliosis progression tracking.
Main Methods:
- A deep learning model was trained on 227 spine radiographs and validated on 70 images.
- The model automatically identifies vertebrae and calculates the Cobb angle.
- Performance was evaluated using absolute error and intraclass correlation coefficient (ICC).
Main Results:
- The model achieved an average absolute error of 1.97° (SD 1.57°).
- 95.9% of measurements had an absolute error under 5°.
- An excellent ICC of 0.981 demonstrated high reliability.
Conclusions:
- The developed deep learning model offers a reliable and potentially valuable tool for clinical scoliosis assessment.
- Automation of Cobb angle measurement can alleviate clinician workload.
- Further research is recommended to enhance model accuracy and applicability.
