Applying Machine Learning and Point-Set Registration to Automatically Measure the Severity of Spinal Curvature on Radiographs

• Published in: IEEE journal of translational engineering in health and medicine Vol. 12; pp. 151 - 161
• Main Authors: Wong, Jason, Reformat, Marek, Lou, Edmond
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.01.2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Acceptance; Adolescent; Algorithms; Artificial neural networks; Child; Convolutional neural network; Convolutional neural networks; Diagnostic radiography; Error analysis; Human error; Humans; Image segmentation; Iterative methods; Kyphosis; Machine learning; Measurement methods; Measurement uncertainty; point-set registration; radiograph; Radiographs; Radiography; Scoliosis; Scoliosis - diagnosis; Segmentation; Spinal curvature; Spine - diagnostic imaging; Time measurement; Training; Vertebrae; Workflow; point-set registration; radiograph; machine learning; Convolutional neural network; scoliosis
• ISSN: 2168-2372; 2168-2372
• DOI: 10.1109/JTEHM.2023.3332618

Objective: Measuring the severity of the lateral spinal curvature, or Cobb angle, is critical for monitoring and making treatment decisions for children with adolescent idiopathic scoliosis (AIS). However, manual measurement is time-consuming and subject to human error. Therefore, clinicians seek an automated measurement method to streamline workflow and improve accuracy. This paper reports on a novel machine learning algorithm of cascaded convolutional neural networks (CNN) to measure the Cobb angle on spinal radiographs automatically. Methods: The developed method consisted of spinal column segmentation using a CNN, vertebra localization and segmentation using iterative vertebra body location coupled with another CNN, point-set registration to correct vertebra segmentations, and Cobb angle measurement using the final segmentations. Measurement performance was evaluated with the circular mean absolute error (CMAE) and percentage within clinical acceptance (<inline-formula> <tex-math notation="LaTeX">\le 5^{\circ } </tex-math></inline-formula>) between automatic and manual measurements. Analysis was separated by curve severity to identify any potential systematic biases using independent samples Student's t-tests. Results: The method detected 346 of the 352 manually measured Cobb angles (98%), with a CMAE of 2.8° and 91% of measurements within the 5° clinical acceptance. No statistically significant differences were found between the CMAEs of mild (<inline-formula> <tex-math notation="LaTeX"> < 25^{\circ } </tex-math></inline-formula>), moderate (25°-45°), and severe (<inline-formula> <tex-math notation="LaTeX">\ge 45^{\circ } </tex-math></inline-formula>) groups. The average measurement time per radiograph was 17.7±10.2s, improving upon the estimated average of 30s it takes an experienced rater to measure. Additionally, the algorithm outputs segmentations with the measurement, allowing clinicians to interpret measurement results. Discussion/Conclusion: The developed method measured Cobb angles on radiographs automatically with high accuracy, quick measurement time, and interpretability, suggesting clinical feasibility.