Skeletal Radiology最新文献

Objectives: To compare the accuracy and reliability of 2D and 3D methods for measuring femoral version against an anatomic reference standard using 3D-printed femoral phantoms.

Methods: CT data from three skeletally mature pediatric patients (2 females, 1 male; age 17.8 ± 0.1 years) were used as digital templates for 3D-printed haptic femur models. The anatomic reference standard for femoral version was determined by an established drilling technique. Models underwent CT scanning in neutral position and variations of flexion, internal/external rotation, and abduction/adduction (in increments of 15°, 30°, 45°). Two radiologists measured femoral version on axial datasets using 2D techniques (Murphy, Lee, Reikerås) and a 3D-reconstructed model in a blinded fashion. Mean absolute error (MAE) between measurements and the reference standard and mean absolute differences (MAD) between readers were calculated.

Results: Across all positions in all femurs, MAE was 22.9° ± 23.7° for the Murphy technique, 12.1° ± 8.5° for the Lee technique, 12.2° ± 9.6° for the Reikerås technique, and 2.3° ± 1.3° for the 3D method. For all 2D methods, MAE was greatest with adduction, flexion, and combinations of both. For the 3D method, femur position had no impact on MAE. MAD between readers was 9.7° for the Murphy technique, 4.7° for the Lee technique, 4.4° for the Reikerås technique, and 2.9° for the 3D method.

Conclusions: Femoral version measurements based on 3D reconstructions are more accurate than traditional 2D techniques, more robust to femur positioning, and more consistent between readers. Accurate measurement of femoral version is important for understanding and treating lower extremity deformities in the skeletally mature pediatric population.

Objective: To develop a multimodal ultrasound radiomics model integrating power Doppler (PD) and superb microvascular imaging (SMI) data, combined with clinical baseline characteristics, and to evaluate its predictive performance for active inflammation in rheumatoid arthritis (RA).

Methods: A total of 2,503 PD and 2,503 SMI ultrasound images of metacarpophalangeal and proximal interphalangeal joints were collected from RA patients with active inflammation and non-active controls between December 2022 and March 2024. Single-modality (PD/SMI), multimodal (SMI_PD), and clinical-integrated models were developed. Diagnostic performance was evaluated using receiver operating characteristic (ROC) analysis.

Results: The area under the curve (AUC) values for PD, SMI, multimodal SMI_PD, and clinical-integrated models were 0.76, 0.75, 0.82, and 0.83, respectively.

Conclusion: The multimodal radiomics model demonstrated superior diagnostic performance for RA active inflammation. SMI shows potential as a tool for subclinical inflammation monitoring, while the integration of multimodal ultrasound radiomics with clinical data further enhances clinical utility.

Objective: To analyze the experience at our institutions with liposclerosing myxofibrous tumor (LSMFT) regarding radiographic reporting, pathologic correlation, and follow-up for malignant transformation.

Methods: A retrospective search of imaging and pathology databases at two tertiary institutions (1987-2022) identified cases with LSMFT in the diagnosis or differential diagnosis. Clinical records, imaging reports, and follow-up data were reviewed. Two fellowship-trained musculoskeletal radiologists independently assessed imaging features and two bone tumor pathologists independently reviewed the pathology findings.

Results: Among over 11 million imaging reports, 149 cases included LSMFT in the imaging report differential diagnosis, with pathologic correlation available in 18 patients (8 men, 10 women; median age 41.9 years [range, 16.5-68.3 years]). All patients presented with hip pain. Radiography was performed in all 18 cases, CT in 7 cases, and MRI in 15 cases. On imaging, lesions appeared well-defined with a sclerotic margin (14/18) and were most commonly central (14/18) and intertrochanteric (16/18) in location. Mean diameter was 5.9 cm in greatest dimension (range, 3.2-11.5 cm). Histopathology revealed low-grade benign neoplastic and non-neoplastic lesions in all cases, most commonly fibrous dysplasia (9/18). Two out of 18 cases were found with LSMFT in the pathology report impression. No malignant transformations were observed on follow-up.

Conclusion: When radiography shows findings described as possible LSMFT, the pathology diagnosis generally corresponds to other benign lesions. No malignant transformations were observed in our cohort. In the absence of symptoms or aggressive imaging features, we do not routinely pursue surveillance cross-sectional imaging, biopsy, or surgical management.

Objectives: Lower extremity fractures are prevalent in vulnerable populations leading to significant burdens, which highlight the need for timely and precise diagnosis. Integrating deep learning with imaging findings has shown promising results for enhancing fracture detection accuracy. This study assesses the diagnostic accuracy of AI models using X-ray images to detect ankle and foot fractures and investigates the probable factors influencing their performance.

Materials and methods: A comprehensive search of four databases was done up to September 30th, 2025. Studies that investigated the accuracy of deep learning models for the detection of ankle and foot fractures utilizing X-rays were included. A bivariate random-effects model was used to perform meta-analysis.

Results: A total of 506 studies were reviewed, and 14 were included in the meta-analysis. Analysis of all the representative models of the included studies had a cumulative sensitivity and specificity of 93.2% and 94.5% (95% CI 88.8-95.9%, 90.1-97.0%, respectively, I² 6.6%). The pooled F1 score was estimated at 0.94 (95% CI 0.88-0.97). Subgroup analysis revealed no difference in the sensitivity or specificity of studies using multi-view vs. single-view X-rays (P = 0.64, P = 0.89, respectively). Models detecting calcaneal fractures performed significantly better than models detecting foot or ankle fractures, with a pooled sensitivity of 95.1% (95% CI 93.0-96.6%, P = 0.008), specificity of 98.3% (95% CI 97.00-99.0%), and a DOR of 1751.8 (95% CI 445.9-6882.5). The type of dataset used (validation vs. test, internal testing vs. external testing) did not significantly affect performance.

Conclusion: Deep learning models can be considered a notably accurate method for the detection of ankle and foot fractures. Further studies with larger samples, external validation, and clinical implementations are required.

Prospero registration: This systematic review and meta-analysis study was registered with PROSPERO, registration number CRD42024624044.

Background: Spinal and sacroiliac disorders, such as disc herniation, sacroiliitis, and spondylarthritis, require precise imaging for diagnosis. Conventional computed tomography (CT) is widely regarded as the reference standard for evaluating cortical bone and certain structural lesions due to its superior spatial resolution and accurate Hounsfield unit quantification, whereas MRI remains superior for marrow edema, soft tissue involvement, and early inflammatory changes. MRI-based synthetic CT (sCT) offers a radiation-free alternative by generating CT-like images from MRI data. This systematic review evaluates the diagnostic accuracy of sCT compared to conventional CT in these disorders.

Methods: Following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) for diagnostic test accuracy studies (PRISMA-DTA) guidelines, we searched databases (PubMed, Embase, Cochrane, Scopus, Web of Science, CINAHL, IEEE Xplore) and gray literature. Eligible studies compared sCT to CT in patients with spinal/sacroiliac disorders, reporting metrics like sensitivity, specificity, and agreement. Two reviewers screened records, extracted data, and assessed risk of bias using QUADAS-2. Narrative synthesis was performed due to heterogeneity.

Results: Ten studies (n = 1028 participants) were included, focusing on degenerative spinal conditions and axial spondyloarthritis. sCT showed sensitivity of 47-94% and specificity of 83->95% for lesions (e.g., erosions, ankylosis), often superior to MRI and comparable to CT. Geometric measurements (e.g., vertebral dimensions) had high concordance (ICC = 0.788-0.978). Inter-observer agreement was moderate to excellent (κ = 0.41-0.98). Risk of bias was low in four studies but high in six, mainly due to small samples and selection issues.

Conclusions: MRI-based sCT demonstrates promising diagnostic accuracy for spinal and sacroiliac disorders, potentially reducing radiation exposure. However, methodological limitations warrant larger, multicenter trials for validation.

Objectives: This paper aims to establish an X-ray imaging grading for assessing ankle joints in adult Kashin-Beck disease (KBD) and investigate its correlation with clinical grading of finger and ankle pain.

Methods: The study involved 160 adult KBD patients (a total of 320 ankles) as the case group and 170 matched healthy subjects (a total of 340 ankles) as the control group. Digital radiographs of bilateral ankle joints were obtained, and the talus trochlea angle was measured according to the radiographs. The X-ray grading of adult KBD ankles was conducted using the Kellgren-Lawrence grading as a reference. Correlation analysis was performed to study the relationship between X-ray grading and clinical grading of the finger.

Result: The X-ray imaging of KBD ankle joints was categorized into grades 0-IV, with grade IV further divided into subtypes a, b, c, and d. There was a weak correlation between ankle X-ray grading and the clinical grading of the fingers in patients with KBD (r = 0.208, p = 0.001). There was a significant correlation between X-ray grading and ankle pain (r = 0.610, p = 0.000), but almost no correlation between ankle pain and the clinical grading of the finger (r = 0.101, p = 0.071).

Conclusion: This study introduced an X-ray grading method for KBD ankles, although it showed weak correlation with the clinical grading of the fingers. Importantly, a moderate correlation was identified between X-ray grading and ankle pain, but almost no significant link was established between ankle pain and the clinical grading of the finger.

Objective: To estimate hip fracture (Fx) risk if one or more vertebral fractures are detected in an otherwise healthy older woman or man.

Materials and methods: The study was conducted among older Chinese community subjects. 2000 women were enrolled; 1951 were followed for five years. 2000 men were enrolled; 1882 cases were followed for five years. With thoracolumbar spine radiographs, for each vertebra, a score of 0, -0.5, -1, -1.5, -2, -2.5, and -3 was assigned for no osteoporotic-like vertebral fractural deformity (OLVF) or OLVF of < 20%, ≥ 20 ~ 25%, ≥ 25% ~ 33%, ≥ 33% ~ 40%, ≥ 40% ~ 66%, and ≥ 66% vertebral height loss, respectively. OLVFss was the summed score of vertebrae T3 to L5. The femoral neck (FN) cutpoint T-score was ≤ -2.7 for osteoporosis in women and ≤ -2.1 for osteofrailia in men. OLVFss ≤  - 1.5 suggests osteoporosis in women, and OLVFss ≤  - 2.5 suggests osteofrailia in men.

Results: For women, 26 cases developed hip Fx (mean age: 80 years). Baseline FN T-score ≤ -2.7 and OLVFss ≤ -1.5 had a positive predictive value of 5.22% and 5.05%, and a detection sensitivity of 65.38% and 53.85%, respectively. For men, 23 cases developed hip Fx (mean age: 81 years). Baseline FN T-score ≤ -2.1 and OVLFss ≤ -2.5 had a positive predictive value of 4.24% and 4.81%, and a detection sensitivity of 52.17% and 21.74%, respectively.

Conclusion: Compared with FN T-score thresholds, OVLFss had a similar performance for future hip Fx risk estimation for women, for men OVLFss had a similar positive predictive value for future hip Fx risk but was associated with a lower detection sensitivity.