Related Experiment Video
Updated: Sep 17, 2025
07:56
Scanning Skeletal Remains for Bone Mineral Density in Forensic Contexts
Published on: January 29, 2018
17.7K
Radiomics based on BPX images to detect abnormal bone mineral density
Yuqi Zhu1, Kun Zhou2, Xiao Luo2
1Department of Radiology, Huashan Hospital, Fudan University, Shanghai, China.
Bone
|July 3, 2025
Summary
Radiomics analysis of biplanar X-ray radiography (BPX) images can accurately identify abnormal bone mineral density (BMD). This automated tool shows promise for risk assessment in clinical settings, reducing the need for radiation-intensive methods.
Area of Science:
- Medical Imaging
- Radiology
- Artificial Intelligence in Medicine
Background:
- Abnormal bone mineral density (BMD) increases fracture risk.
- Current diagnostic methods like DXA and QCT involve radiation exposure and costs.
- Investigating radiomics as a low-cost, radiation-free alternative for BMD assessment.
Purpose of the Study:
- To assess the feasibility of using radiomics on biplanar X-ray radiography (BPX) images.
- To develop an automated tool for identifying patients with high risk of BMD abnormality.
- To compare radiomics-based models with traditional methods.
Main Methods:
- Utilized 906 BPX scans from 453 subjects, with QCT as the ground truth.
- Extracted radiomic features from L1-L5 vertebrae using Pyradiomics.
- Developed radiomics-only and clinical-radiomics models, employing LASSO for feature selection.
Main Results:
- The radiomics-only model achieved 0.88 accuracy for both sexes and 0.91 for women.
- Clinical-radiomics models showed 0.87 accuracy for both sexes and 0.89 for women.
- Radiomics models demonstrated high performance in diagnosing BMD abnormality.
Conclusions:
- A radiomics-only model using BPX images effectively identifies BMD abnormalities.
- This approach shows potential as a decision-support tool in routine physical examinations.
- Radiomics offers a promising, radiation-free method for BMD risk stratification.
