Comparative efficacy of anteroposterior and lateral X-ray based deep learning in the detection of osteoporotic vertebral compression fracture

Abstract

Magnetic resonance imaging remains the gold standard for diagnosing osteoporotic vertebral compression fractures (OVCF), but the use of X-ray imaging, particularly anteroposterior (AP) and lateral views, is prevalent due to its accessibility and cost-effectiveness. We aim to assess whether the performance of AP images-based deep learning is comparable compared to those using lateral images. This retrospective study analyzed X-ray images from two tertiary teaching hospitals, involving 1,507 patients for the training and internal test, and 104 patients for the external test. The EfficientNet-B5-based algorithms were employed to classify OVCF and non-OVCF group. The model was trained with a 1:1 balanced dataset and validated through 5-fold cross validation. Performance outcomes were compared with the area under receiver operating characteristic (AUROC) curve. Out of a total of 1,507 patients, 799 were included in the training dataset and 708 were included in the internal test dataset. The training and internal test datasets were matched 1:1 as OVCF and non-OVCF patients. The DL model showed comparable classifying performance with internal test data (N = 708, AUROC for AP, 0.915; AUROC for lateral, 0.953) and external test data (N = 104, AUROC for AP, 0.982; AUROC for lateral, 0979), respectively. The other performances including F1 score and accuracy were also comparable. Especially, The AUROC of AP and lateral x-ray image-based DL was not significantly different (p for DeLong test = 0.604). The EfficientNet-B5 algorithms using AP X-ray images shows comparable efficacy for classifying OVCF and non-OVCF compared to lateral images.

Keywords: Anteroposterior; Deep learning; Diagnostic accuracy; Lateral views; Machine learning in radiology; Osteoporotic vertebral compression fractures; X-ray imaging.

Fig. 1

Typical X-ray images of osteoporotic vertebral compression fracture. The AP image (A) does not evidently show a compression fracture at the L1 level (arrow head), but the lateral image (B) shows wedging and deformity of the spinal column (arrow), which are typical findings of a osteoporotic vertebral compression fracture.

Fig. 2

Flow diagram of the participants. OVCF, osteoporotic vertebral compression fracture; TB, tuberculosis.

Fig. 3

Sample AP X-ray images before and after preprocessing.

Fig. 4

Schematic architecture of the deep learning classification model. means EfficientNetV2-s

Fig. 5

Result of the performance of deep learning classification model of the anteroposterior (a and c) and lateral (b and d) X-ray images in the internal (a and b) and external (c and d) test dataset. Label 0, non-osteoporotic vertebral compression fracture; label 1, osteoporotic vertebral compression fracture.

Fig. 6

Comparison of the AUROC curve for the deep learning model to classify OVCF and non-OVCF using anteroposterior and lateral X-ray images in the internal (a) and external (b) test dataset. OVCF; osteoporotic vertebral compression fracture.