CT-Based Deep Learning for Differentiating Acute and Chronic Osteoporotic Vertebral Compression Fractures (CT-DL-OVCF)

A Deep Learning Model Based on CT Images for Differentiating Acute and Chronic Osteoporotic Vertebral Compression Fractures

Osteoporotic vertebral compression fractures are common in older adults and may present as either acute or chronic fractures. Correctly distinguishing acute from chronic fractures is clinically important because treatment strategies and management decisions differ depending on fracture chronicity. However, differentiating acute and chronic osteoporotic vertebral compression fractures based on imaging findings alone can be challenging in routine clinical practice.

This retrospective study aims to develop an intelligent diagnostic system based on computed tomography (CT) images to differentiate acute and chronic osteoporotic vertebral compression fractures. Clinical and imaging data from patients diagnosed with osteoporotic vertebral compression fractures will be collected from the First Affiliated Hospital of Chongqing Medical University and an additional medical center. A deep learning model will be trained to automatically analyze CT images and classify fractures as acute or chronic.

The results of this study may help improve the accuracy and efficiency of fracture chronicity assessment using CT images and provide supportive information for clinical decision-making regarding treatment selection in patients with osteoporotic vertebral compression fractures.

Study Overview

• Status: Not yet recruiting
• Conditions: Osteoporotic Vertebral Compression Fractures
• Intervention / Treatment: Other: No Intervention (Observational Study)

Detailed Description

This study is a retrospective, multicenter observational study designed to develop and evaluate a deep learning-based system for differentiating acute and chronic osteoporotic vertebral compression fractures using computed tomography (CT) images.

Patients diagnosed with osteoporotic vertebral compression fractures who underwent both CT and magnetic resonance imaging (MRI) examinations will be retrospectively collected from the First Affiliated Hospital of Chongqing Medical University and one additional medical center between January 2023 and September 2025. Clinical data, including age, sex, and dual-energy X-ray absorptiometry (DXA) results, as well as complete DICOM-format CT and MRI images, will be collected. The interval between CT and MRI examinations must be less than two weeks. Patients with pathological fractures caused by infection or tumor, the presence of foreign materials such as bone cement or metallic hardware, or poor image quality with significant artifacts will be excluded.

The study workflow includes data collection, model development, performance evaluation, and model interpretability analysis. Multiple deep learning segmentation models, including U-Net, U-Mamba, and UNETR++, will first be evaluated for vertebral body segmentation performance. Based on the optimal segmentation results, classification models such as VGG-16, DenseNet-121, Vision Transformer (ViT), and Transformer-based architectures will be trained to differentiate acute and chronic compression fractures. The best-performing model will be selected to construct the final classification system.

Model performance for segmentation tasks will be assessed using Dice similarity coefficient and loss values. Classification performance will be evaluated in an external validation dataset using area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, positive predictive value, and negative predictive value. Receiver operating characteristic curves and confusion matrices will be generated to visualize model performance.

To improve model interpretability, gradient-weighted class activation mapping (Grad-CAM) will be applied to generate heatmaps highlighting image regions that contribute most to model predictions. These heatmaps will be overlaid on CT images to visually demonstrate how the model differentiates acute and chronic osteoporotic vertebral compression fractures.

Based on a predefined sample size calculation assuming a sensitivity of 0.90, a significance level of 0.05, and an allowable error of 0.05, a total of 276 patients (138 acute and 138 chronic cases) are expected to be included in this study.

• Study Type: Observational
• Enrollment (Estimated): 276

Contacts and Locations

Study Locations

• China
  • Chongqing Municipality
    • Chongqing, Chongqing Municipality, China, 400016
      • The First Affiliated Hospital of Chongqing Medical University
      • Contact: Xin Fan; Phone Number: +86 23 89011876; Email: 202770@hospital.cqmu.edu.cn
      • Contact: Email: 202770@hospital.cqmu.edu.cn

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

The study population consists of adult patients aged 40 years or older who were diagnosed with osteoporotic vertebral compression fractures and underwent CT and MRI examinations at the participating centers.

Description

Inclusion Criteria:

• Inclusion Criteria:
• Patients diagnosed with osteoporotic vertebral compression fractures.
• Patients who underwent both CT and MRI examinations of the spine, with an interval of less than 2 weeks between examinations.
• Availability of complete CT and MRI imaging data in DICOM format.
• Availability of complete clinical information, including age, sex, and dual-energy X-ray absorptiometry (DXA) results.
• Age 50 years or older at the time of imaging.

Exclusion Criteria:

• Vertebral compression fractures caused by infection or malignancy.
• Presence of foreign materials, including bone cement or metallic hardware.
• Poor image quality or significant imaging artifacts that affect analysis.

Study Plan

How is the study designed?

Number of groups / cohorts

2

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Acute Osteoporotic Vertebral Compression Fracture Group; Patients diagnosed with acute osteoporotic vertebral compression fractures based on clinical assessment and imaging findings.	Other: No Intervention (Observational Study); This is a retrospective observational study. No therapeutic, diagnostic, or preventive intervention is assigned as part of the study. All analyses are based on previously acquired clinical and imaging data.
Chronic Osteoporotic Vertebral Compression Fracture Group; Patients diagnosed with chronic osteoporotic vertebral compression fractures based on clinical assessment and imaging findings.	Other: No Intervention (Observational Study); This is a retrospective observational study. No therapeutic, diagnostic, or preventive intervention is assigned as part of the study. All analyses are based on previously acquired clinical and imaging data.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Diagnostic performance of the deep learning model for differentiating acute and chronic osteoporotic vertebral compression fractures	The diagnostic performance of the deep learning model in differentiating acute and chronic osteoporotic vertebral compression fractures based on CT images, evaluated using the area under the receiver operating characteristic curve (AUC).	At the time of image analysis

Collaborators and Investigators

• Sponsor: Xin Fan

Study record dates

Study Major Dates

• Study Start (Estimated): December 16, 2025
• Primary Completion (Estimated): December 29, 2025
• Study Completion (Estimated): January 15, 2026

Study Registration Dates

• First Submitted: December 15, 2025
• First Submitted That Met QC Criteria: December 15, 2025
• First Posted (Estimated): December 29, 2025

Study Record Updates

• Last Update Posted (Actual): December 31, 2025
• Last Update Submitted That Met QC Criteria: December 26, 2025
• Last Verified: December 1, 2025

More Information

Terms related to this study

• Keywords: Deep learning; Osteoporotic vertebral compression fracture
• Additional Relevant MeSH Terms: Investigative Techniques; Methods; Observation
• Other Study ID Numbers: KX2025-KYC1056-01

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: UNDECIDED
• IPD Plan Description: A final decision on sharing individual participant data has not been made at the time of registration. Potential data sharing will be considered in accordance with institutional review board approval, patient privacy protection, and relevant data governance policies.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No