A Super-Resolution Diffusion Model for Recovering Bone Microstructure from CT Images
- 1 November 2023
- journal article
- research article
- Published by Radiological Society of North America (RSNA) in Radiology: Artificial Intelligence
- Vol. 5 (6) , e220251
Abstract
To use a diffusion-based deep learning model to recover bone microstructure from low-resolution images of the proximal femur, a common site of traumatic osteoporotic fractures. Training and testing data in this retrospective study consisted of high-resolution cadaveric micro-CT scans (n = 26), which served as ground truth. The images were downsampled prior to use for model training. The model was used to increase spatial resolution in these low-resolution images threefold, from 0.72 mm to 0.24 mm, sufficient to visualize bone microstructure. Model performance was validated using microstructural metrics and finite element simulation–derived stiffness of trabecular regions. Performance was also evaluated across a handful of image quality assessment metrics. Correlations between model performance and ground truth were assessed using intraclass correlation coefficients (ICCs) and Pearson correlation coefficients. Compared with popular deep learning baselines, the proposed model exhibited greater accuracy (mean ICC of proposed model, 0.92 vs ICC of next best method, 0.83) and lower bias (mean difference in means, 3.80% vs 10.00%, respectively) across the physiologic metrics. Two gradient-based image quality metrics strongly correlated with accuracy across structural and mechanical criteria (r > 0.89). The proposed method may enable accurate measurements of bone structure and strength with a radiation dose on par with current clinical imaging protocols, improving the viability of clinical CT for assessing bone health. Keywords: CT, Image Postprocessing, Skeletal-Appendicular, Long Bones, Radiation Effects, Quantification, Prognosis, Semisupervised Learning Online supplemental material is available for this article. © RSNA, 2023Keywords
Funding Information
- National Science Foundation (2026906)
- National Institutes of Health (R01 AR068282, R01 AR076392)
This publication has 30 references indexed in Scilit:
- Computationally-Optimized Bone Mechanical Modeling from High-Resolution Structural ImagesPLOS ONE, 2012
- High-resolution Computed Tomography for Clinical Imaging of Bone MicroarchitectureClinical Orthopaedics and Related Research, 2011
- BoneJ: Free and extensible bone image analysis in ImageJBone, 2010
- Computational biomechanics of the distal tibia from high-resolution MR and micro-CT imagesBone, 2010
- Age- and gender-dependent changes in three-dimensional microstructure of cortical and trabecular bone at the human femoral neckOsteoporosis International, 2009
- Vertebral Osteoporosis and Trabecular Bone QualityAnnals of Biomedical Engineering, 2006
- Bone Quality — The Material and Structural Basis of Bone Strength and FragilityNew England Journal of Medicine, 2006
- Epidemiology and outcomes of osteoporotic fracturesThe Lancet, 2002
- Trabecular Bone Microarchitecture, Bone Mineral Density, and Vertebral Fractures in Male OsteoporosisJournal of Bone and Mineral Research, 2000
- A Comparative Study of Trabecular Bone Properties in the Spine and Femur Using High Resolution MRI and CTJournal of Bone and Mineral Research, 1998