Deep learning algorithm for the automatic assessment of axial vertebral rotation in patients with scoliosis using the Nash-Moe method

Abstract

Accurate assessments of axial vertebral rotation (AVR) is essential for managing idiopathic scoliosis. The Nash-Moe classification method has been extensively used for AVR assessment; however, its subjective nature can lead to measurement variability. Therefore, herein, we propose an automated deep learning (DL) model for AVR assessment based on posteroanterior spinal radiographs. We develop a two-stage DL framework using the MMRotate toolbox and analyze 1080 posteroanterior spinal radiographs of patients aged 4-18 years. The framework comprises a vertebra detection model (864 training and 216 validation images) and a pedicle detection model (14,608 training and 3652 validation images). We improved the Nash-Moe classification method by implementing a 12-segment division system and width ratio metric for precise pedicle assessment. The vertebra and pedicle detection models achieved mean average precision values of 0.909 and 0.905, respectively. The overall classification accuracy was 0.74, with grade-specific performance between 0.70 and 1.00 for precision and 0.33 and 0.93 for recall across Grades 0-3. The proposed DL framework processed complete posteroanterior radiographs in < 5 s per case compared with conventional manual measurements (114 s per radiograph). The best performance was observed in mild to moderate rotation cases, with performance in severe rotation cases limited by insufficient data. The implementation of DL framework for the automated Nash-Moe classification method exhibited satisfactory accuracy and exceptional efficiency. However, this study is limited by low recall (0.33) for Grade 3 and the inability to classify Grade 4 towing to dataset constraints. Further validation using augmented datasets that include severe rotation cases is necessary.

Keywords: Deep learning; Radiography; Rotation; Scoliosis; Spine.

Fig. 1

Workflow of automated vertebra and pedicle detection with subsequent Nash–Moe grade classification.

Fig. 2

Illustration of Nash–Moe classification method for assessing axial vertebral rotation.

Fig. 3

Schematic of single vertebra divided into 12 equal regions, including pedicle positioning.

Fig. 4

Illustration of pedicle coordinate positions identified at segments (2, 12) on vertebral division grid.

Fig. 5

(Left) Example of vertebra annotation and (Right) predicted vertebra bounding boxes generated by trained object detection model.

Fig. 6

Confusion matrix displaying Nash–Moe grade classification results.