TL;DR
DL analysis of lumbar radiographs, combined with transfer learning and metaheuristic optimization, may support BMD category classification as a complementary screening tool for osteoporosis screening and diagnosis.
OBJECTIVE To develop and validate a hierarchical deep learning model for differentiating acute and chronic lumbar osteoporotic vertebral compression fractures (OVCFs) using X-ray images. MATERIALS AND METHODS We retrospectively reviewed approximately 2600 lateral lumbar radiographs obtained from patients clinically suspected of having OVCFs between 2007 and 2022. After excluding poor-quality images and surgically instrumented vertebrae, 1299 radiographs (6495 vertebral patches, L1-L5) were included. Labeling was performed by neurosurgeons and radiologists using X-ray images, with CT and/or MRI findings serving as the reference standard. A two-step hierarchical classification was implemented: first classifying vertebrae into Normal-Chronic, Acute, and Indeterminate (cement-augmented vertebrae without instrumentation) groups, followed by subdivision of the Normal-Chronic group into Normal and Chronic categories. RESULTS A total of 1299 radiographs were evaluated. The hierarchical model achieved an accuracy of 91% in the initial three-class step. For the detection of acute fractures in the final classification step, the model demonstrated a sensitivity of 91.0% (95% CI 84.8-95.0%), a specificity of 82.1% (95% CI 79.8-84.5%), and a high negative predictive value (NPV) of 98.8% (95% CI 97.9-99.3%). The Normal-aligned hierarchical approach outperformed the Acute-aligned and end-to-end models, particularly for acute and chronic cases. CONCLUSION The proposed hierarchical approach enhances the diagnostic utility of standard X-ray images by enabling more accurate classification of lumbar fracture types. This study is limited by its single-institution retrospective design. This model may reduce reliance on advanced imaging and support faster and more informed clinical decision-making.
BACKGROUND The gold standard for low bone mineral density (BMD) screening is dual-energy X-ray absorptiometry (DXA) scanning, which has declined in use in recent years for a variety of reasons. This decline warrants alternative screening mechanisms. This study aimed to determine if opportunistic foot radiographs can serve as a screening tool for predicting low BMD as measured by DXA scan. METHODS The fifth metatarsal index (5MI), a novel measurement made using the oblique foot radiograph to evaluate cortical thickness, was calculated for patients 50 years and older who underwent a DXA scan and a foot radiograph within 1 year. T scores and BMD were documented for each patient. Receiver operating characteristic curves were generated, and areas under the curves were calculated to determine which 5MI had optimized sensitivity and specificity for identifying osteoporosis. RESULTS A total of 89 patients were included in the study cohort. The average 5MI in patients with normal BMD, osteopenia, and osteoporosis was 52.6%, 44.6%, and 35.2%, respectively. The optimal 5MI for identifying osteoporosis based on the DXA-derived T scores was 46.8%, which correlated with an area under the curve of 0.977. This optimal cutoff for identifying osteoporosis had a sensitivity of 93.3% (95% CI, 70.2%-99.7%) and a specificity of 95.0% (95% CI, 76.4%-99.7%) (P < .001). CONCLUSION This study describes a potential screening method using low-cost foot radiographs to identify patients with low BMD, which may support earlier detection and referral for bone health evaluation. LEVEL OF EVIDENCE Level III, retrospective case series.
Both HU and VBQ offer useful assessments of vertebral bone quality, but HU provides higher and more stable diagnostic accuracy, demonstrating higher sensitivity, specificity, and accuracy for detecting osteopenia and osteoporosis.
BACKGROUND Osteoporosis is frequently underdiagnosed in trauma patients. However, contrast-enhanced whole-body computed tomography (WBCT) in polytrauma care enables opportunistic assessment of bone status without additional radiation exposure. The aim of this study was to investigate whether vertebral, femoral, and dentomaxillofacial CT parameters can identify patients with documented osteopenia or osteoporosis and whether a combined score improves classification. MATERIALS AND METHODS This retrospective, single-center study included 60 WBCT cases. Twenty patients with osteopenia or osteoporosis documented in their initial trauma report were compared with 20 age- and sex-matched reference patients and 20 younger patients. The final classification of bone status was based on an evaluation of patient records, including dual-energy X‑ray absorptiometry (DXA) data (where available), and medical history. Trabecular density was measured in the lumbar spine (L1-L3), femoral neck, and maxillary alveolar ridge. Hard palate cortical thickness (HPT) was measured in millimeters. The whole-body osteoporosis screening score was defined as WBOS = ƒ(HPT, hip HU, lumbar HU). RESULTS Measurements of HPT, maxilla, hip, and lumbar spine decreased significantly in patients with osteopenia/osteoporosis patients. The combined WBOS score demonstrated the best exploratory performance, with a sensitivity of 100.0%, a specificity of 97.1%, an accuracy of 98.3%, and an AUC of 0.994 at a preliminary threshold of 361. CONCLUSION Opportunistic multisite assessment of bone status using WBCT in polytrauma appears feasible. The WBOS showed the best exploratory discrimination but requires further validation against DXA or quantitative CT before clinical implementation.
Introduction: Osteoporosis is a systemic skeletal disorder that increases the risk of fragility fracture and constitutes a major global health burden. Dual-Energy X-Ray Absorptiometry (DEXA) is the clinical reference standard for Bone Mineral Density (BMD) assessment but, as a two-dimensional projection technique, does not directly evaluate trabecular microarchitecture or marrow composition. Magnetic Resonance Imaging (MRI) signal characteristics of vertebral marrow have been proposed as opportunistic markers of Vertebral Bone Quality (VBQ). Aim: To evaluate the correlation between lumbar spine MRI signal alterations (T1- and T2-weighted) and DEXA-derived BMD parameters, including BMD, T-score and Z-score, in patients undergoing lumbar spine imaging. Materials and Methods: A cross-sectional analytical study was conducted at Assam Down Town University, Guwahati, Assam, India, from October 2023 to April 2024. A total of 100 adults of either sex who underwent both lumbar spine MRI (1.5 Tesla) and DEXA were evaluated. A circular Region Of Interest (ROI) of 1 cm² was placed in the central trabecular portion of the L3 vertebral body on T1- and T2- weighted sagittal images, avoiding cortical bone, end-plate changes and focal lesions. DEXA-derived BMD was classified according to World Health Organisation (WHO) T-score criteria. Pearson’s correlation coefficient with 95% Confidence Intervals (CI) was used to assess associations; a two-tailed p-value <0.05 was considered significant. Results: The cohort comprised 52 females (52%) and 48 males (48%). Based on lumbar (L1-L4) T-scores, 42 (42%) participants had normal BMD, 31 (31%) osteopenia and 27 (27%) osteoporosis. Reduced BMD was more frequent among females and older participants. The T-score showed weak but statistically significant negative correlations with T1-weighted (r=-0.214; 95% CI -0.394 to -0.018; p-value=0.033) and T2- weighted (r=-0.208; 95% CI -0.389 to -0.012; p-value=0.038) signal values. The Z-score correlated strongly with mean BMD (r=0.685; 95% CI 0.565 to 0.777; p-value <0.001); as the Z-score is derived from BMD, this served as an internal consistency check. Conclusion: Lumbar vertebral MRI signal alterations showed weak but statistically significant associations with DEXAderived BMD, suggesting that routinely acquired lumbar MRI may provide supplementary, opportunistic information on VBQ. The modest strength of these correlations indicates that MRI signal cannot currently replace DEXA for the diagnosis of osteoporosis.
A two-stage decomposition that separates systemic skeletal fragility from within-patient vertebral outlier status produces well-calibrated per-vertebra fracture-risk estimates from routine clinical lumbar spine CT and opportunistic abdominal CT.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
