What question did this study set out to answer?

To develop and evaluate a multimodal prediction model that integrates X-ray imaging, deep learning, radiomics, and clinical variables to identify the stage of vertebral compression fractures.

August 22, 2026European Journal of RadiologyOpen Access

Automated stage classification of vertebral compression fractures using multimodal X-ray image analysis

View Full Paper

Ask AI

Bookmark

Authors

SDShenyang DuanNingbo University YDYufeng DengNingbo University QPQingjiang PangNingbo No. 2 Hospital

Discussion

Member takes

Implication

Diagnostic study demonstrates accurate staging of vertebral compression fractures from multimodal X-ray data, highlighting its utility for clinical decision support.

Key Points

To develop and evaluate a multimodal prediction model that integrates X-ray imaging, deep learning, radiomics, and clinical variables to identify the stage of vertebral compression fractures.
Analyzed X-ray images from 2,129 patients (2,787 MRI-confirmed fractured vertebrae) from Ningbo Second Hospital, split into training, validation, and test cohorts in an 8:1:1 ratio.
Extracted deep learning features using pre-trained ResNet34, ResNet50, VGG16, and VGG19 networks alongside radiomics features extracted via PyRadiomics.
Fused imaging features with patient clinical data and evaluated classification performance across five machine learning algorithms: logistic regression, random forest, SVM, XGBoost, and LightGBM.
The combination of VGG16 deep feature extraction, radiomics, clinical data, and logistic regression yielded the highest diagnostic performance.
The top-performing multimodal model achieved an overall accuracy of 0.986 and an area under the receiver operating characteristic curve of 0.988 (95% CI: 0.975–0.990).
Decision curve analysis confirmed significant net clinical benefit across the majority of threshold probabilities.

Cite This Study

Duan et al. (2026) studied this question.

synapsesocial.com/papers/6a9f796cf5b42081939655f7 https://doi.org/10.1016/j.ejrad.2026.113180

View Full Paper

Ask AI

Bookmark

Implication

Diagnostic study demonstrates accurate staging of vertebral compression fractures from multimodal X-ray data, highlighting its utility for clinical decision support.

Automated stage classification of vertebral compression fractures using multimodal X-ray image analysis

Discussion

Member takes

Implication

Key Points

Cite This Study

Automated stage classification of vertebral compression fractures using multimodal X-ray image analysis

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider