Skeletal Radiology最新文献

{"title":"Applications and performance of imaging artificial intelligence in detecting and staging avascular necrosis of the femoral head: a systematic review and meta-analysis.","authors":"Peyman Mirghaderi, Nasim Eshraghi, Hyun Su Kim, Chankue Park, Navid Faraji, Majid Chalian, Peter C Thurlow","doi":"10.1007/s00256-026-05376-z","DOIUrl":"https://doi.org/10.1007/s00256-026-05376-z","url":null,"abstract":"<p><strong>Objective: </strong>To evaluate the diagnostic accuracy of imaging artificial intelligence (AI) models for detecting avascular necrosis of the femoral head (AVNFH) and to examine whether diagnostic performance differs according to imaging modality, disease-stage focus, and validation strategy.</p><p><strong>Materials and methods: </strong>In accordance with PRISMA-DTA, we conducted a systematic search in four main databases. Two independent reviewers evaluated the studies, extracted relevant data, and assessed the quality following the QUADAS-2 criteria. The primary analysis pooled overall sensitivity and specificity using a bivariate random-effects model. Prespecified subgroup analyses assessed differences by imaging modality, disease stage focus, and evaluation strategy.</p><p><strong>Results: </strong>Nine selected estimates contributed to the primary analysis of binary AVNFH detection. Pooled sensitivity was 0.888 (95% CI, 0.789-0.944), specificity was 0.937 (95% CI, 0.906-0.958), and SROC AUC was 0.967. MRI-based estimates showed higher sensitivity than radiography-based estimates in the study-level comparison (p = 0.027). No significant subgroup differences were found by disease-stage focus and validation strategy (all p > 0.05). Only five external test-set estimates contributed to the primary analysis; subgroup comparisons had limited precision.</p><p><strong>Conclusion: </strong>Imaging AI models showed promising accuracy for AVNFH detection in research datasets. However, risk of bias, between-study heterogeneity, and limited external testing restrict confidence in their clinical applicability. Prospective, multicenter validation and evaluation of clinical benefit are needed before routine implementation.</p>","PeriodicalId":21783,"journal":{"name":"Skeletal Radiology","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149115125","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Regional variation in vertebral bone marrow adiposity: a population-based atlas of age- and sex-specific patterns.","authors":"Ulf Bach, Thomas Uslar, Charlotte Fünning, Guney Tombak, Katharina Frenes, Joan Walter, Philipp A Gerber, Felix Beuschlein, Ender Konukoglu, Thomas Frauenfelder, Ertunc Erdil, Falko Ensle","doi":"10.1007/s00256-026-05381-2","DOIUrl":"https://doi.org/10.1007/s00256-026-05381-2","url":null,"abstract":"<p><strong>Background: </strong>Bone marrow adipose tissue (BMAT) is a metabolically active fat depot implicated in skeletal aging and systemic metabolism. While chemical shift-encoding MRI enables noninvasive quantification of bone marrow fat fraction (BMFF), current knowledge of vertebral BMAT is largely limited to small cohorts or restricted anatomical regions.</p><p><strong>Purpose: </strong>To characterize the spatial distribution of vertebral bone marrow fat fraction across the spine and determine region-specific associations with age and sex in a population-based cohort.</p><p><strong>Materials and methods: </strong>In this retrospective study, MRI data from 82,920 UK Biobank participants recruited between 2006 and 2010 were analyzed. The baseline case subset (n = 81,318) had a mean age of 54.8 ± 7.6 years, and 42,113/81,318 (51.8%) were female. Automated vertebral body segmentation enabled voxel-wise BMFF quantification using Dixon MRI. Vertebra- and region-level BMFF metrics were derived. Associations with age and sex were assessed using region-specific linear regression.</p><p><strong>Results: </strong>Vertebral BMAT demonstrated a non-linear cranio-caudal pattern. Mean BMFF was ~ 39% in the cervical, ~ 55% in the thoracic, ~ 45% in the lumbar spine, and ~ 70% in the sacrum. BMAT increased with age in all regions (all p < 0.001; increase per decade: cervical 3.33%, thoracic 3.41%, lumbar 2.95%, sacral 3.83%). Females exhibited higher BMFF than males across all regions (all p < 0.001). The overall spatial pattern remained consistent across age groups and sexes.</p><p><strong>Conclusion: </strong>Vertebral BMAT exhibits a structured, non-linear spatial distribution with region-specific age- and sex-dependent patterns. This population-based atlas provides normative reference data supporting region-specific BMAT assessment in skeletal and metabolic disease.</p>","PeriodicalId":21783,"journal":{"name":"Skeletal Radiology","volume":" ","pages":""},"PeriodicalIF":2.5,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149115290","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Discrepancies in CPAK classification between CT and long-leg radiography: a systematic review and meta-analysis.","authors":"Tao Bian, Yunfeng Zhang, Lei Li, Yixin Zhou","doi":"10.1007/s00256-026-05224-0","DOIUrl":"10.1007/s00256-026-05224-0","url":null,"abstract":"<p><strong>Objective: </strong>To determine whether substantial differences in coronal plane alignment of the knee phenotype distribution, as well as systematic angular measurement discrepancies, exist between CT and long-leg radiography.</p><p><strong>Materials and methods: </strong>From February 2021 to April 2025, we searched PubMed, Embase, and the Cochrane Central Register of Controlled Trials for studies comparing CT- and long-leg radiography-derived coronal plane alignment classifications of the knee in patients with osteoarthritis. The primary outcome was distribution of coronal plane alignment phenotypes. Secondary outcomes included differences in medial proximal tibial angle, lateral distal femoral angle, arithmetic hip-knee-ankle angle, and joint line obliquity.</p><p><strong>Results: </strong>Four studies (1,134 knees) were included. Compared with long-leg radiography-derived classification, CT-derived classification increased type I phenotypes (risk difference: 0.10; 95% confidence interval: 0.01-0.20; P = 0.040) and decreased type III (risk difference: -0.04; 95% confidence interval: -0.07 to -0.01; P = 0.020) and type V phenotypes (risk difference: -0.04; 95% confidence interval: -0.07 to -0.01; P = 0.004). CT yielded significantly lower medial proximal tibial angle (weighted mean difference: - 1.18°; P < 0.001), arithmetic hip-knee-ankle angle (weighted mean difference: - 0.95°; P < 0.001), and joint line obliquity (weighted mean difference: - 1.40°; P < 0.001) than long-leg radiography. Heterogeneity was high for type I phenotype (I² = 81%), lateral distal femoral angle (I² = 70%), and joint line obliquity (I² = 69%).</p><p><strong>Conclusion: </strong>Discrepancies between CT-based software-generated and long-leg radiography-derived measurements substantially affect coronal plane alignment classification and angular parameters. Surgeons should consider these modality-specific variations and employ compensatory verification strategies to ensure optimal alignment.</p>","PeriodicalId":21783,"journal":{"name":"Skeletal Radiology","volume":" ","pages":"2447-2456"},"PeriodicalIF":2.5,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147820589","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Subperiosteal abscess in pediatric acute osteomyelitis: ultrasound-guided percutaneous aspiration in three cases.","authors":"Gonzalo Corral","doi":"10.1007/s00256-026-05269-1","DOIUrl":"10.1007/s00256-026-05269-1","url":null,"abstract":"<p><p>Subperiosteal abscess is a recognized complication of pediatric acute osteomyelitis and has traditionally been managed with surgical drainage. Evidence supporting minimally invasive ultrasound-guided therapeutic alternatives in children remains limited. We report three pediatric cases of acute osteomyelitis complicated by subperiosteal abscesses successfully treated with ultrasound-guided percutaneous aspiration. In all patients, ultrasound demonstrated subperiosteal collections measuring 4-7 mm in thickness, with no evidence of substantial intraosseous abscess on ultrasound or magnetic resonance imaging. Percutaneous aspiration was performed under real-time ultrasound guidance using fine-caliber needles, without indwelling catheter placement. Purulent material was obtained in all cases, with cultures positive for methicillin-sensitive Staphylococcus aureus. All patients showed rapid clinical and laboratory improvement within 24-48 h, without procedure-related complications, recurrence during follow-up, or subsequent need for surgical drainage. These cases illustrate the feasibility and potential clinical value of ultrasound-guided percutaneous aspiration as a minimally invasive therapeutic option in carefully selected pediatric patients with isolated subperiosteal abscesses.</p>","PeriodicalId":21783,"journal":{"name":"Skeletal Radiology","volume":" ","pages":"2689-2693"},"PeriodicalIF":2.5,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148056581","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Diagnostic accuracy of MRI for meniscal tear patterns in ACL-deficient knees: comparison with arthroscopic findings.","authors":"Artit Boonrod, Arunnit Boonrod, Watcharapong Wongsawiang, Punthip Thammaroj, Sermsak Sumanont, Punyawat Apiwatanakul","doi":"10.1007/s00256-026-05288-y","DOIUrl":"10.1007/s00256-026-05288-y","url":null,"abstract":"<p><strong>Background and objectives: </strong>Anterior cruciate ligament (ACL) injuries are frequently associated with meniscal tears. MRI diagnostic performance varies by tear pattern and location and may be influenced by imaging timing and reader variability. This study evaluated pattern-specific MRI accuracy compared with arthroscopic findings and assessed the impact of timing and inter-reader reliability.</p><p><strong>Methods: </strong>This retrospective study included consecutive patients undergoing ACL reconstruction with confirmed complete ACL tears and available preoperative MRI. Meniscal tears were classified as horizontal, vertical, radial, complex, and bucket-handle. Arthroscopy served as the reference standard. Diagnostic metrics, inter-reader agreement (κ), and time intervals between injury, MRI, and surgery were analyzed.</p><p><strong>Results: </strong>A total of 143 patients were included. Medial meniscal tears were present in 81 patients (56.6%) and lateral tears in 69 (48.3%). MRI accuracy was 71.3% for medial tears (sensitivity 86.4%, specificity 51.6%) and 72.0% for lateral tears (sensitivity 88.4%, specificity 56.8%). Bucket handle tear patterns showed the highest inter-reader reliability (κ = 0.78), while radial tears demonstrated the lowest inter-reader reliability (κ = 0.16). In adjusted analyses, neither meniscal location nor any injury-MRI, MRI-surgery, or injury-surgery interval was significantly associated with diagnostic performance (all p ≥ 0.38).</p><p><strong>Conclusions: </strong>MRI shows moderate accuracy and fair-to-substantial reliability for meniscal tears in ACL-deficient knees, with the best performance in bucket-handle tear patterns. The greatest inter-reader variability was observed for radial tears. Although no significant association between timing intervals and diagnostic performance was identified, timing-related findings should be interpreted cautiously because of potential measurement error and the retrospective design.</p><p><strong>Registration: </strong>The study registration was not performed due to the retrospective design.</p>","PeriodicalId":21783,"journal":{"name":"Skeletal Radiology","volume":" ","pages":"2603-2612"},"PeriodicalIF":2.5,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148340072","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mechanical axis image quality of the EOS imaging system compared to digital radiographs.","authors":"Ahmed Alghamdi, Natalie Lee, Farag Shuweihdi, Amaka C Offiah","doi":"10.1007/s00256-026-05264-6","DOIUrl":"10.1007/s00256-026-05264-6","url":null,"abstract":"<p><strong>Objective: </strong>To compare mechanical axis image quality between the EOS imaging system and digital radiograph (DR).</p><p><strong>Materials and methods: </strong>This retrospective study included 100 patients who underwent both EOS and DR mechanical axis imaging. It was conducted according to the Quality and Standards Department at the University of Sheffield. Three observers independently scored the DR and EOS images according to the modified European guidelines on quality criteria for diagnostic radiographic images in paediatrics: resolution, coverage of the area of interest, positioning, contrast, and motion. The range of scores was (Yes) for the fulfilled criterion and (No) for the not fulfilled criterion. For the motion criterion, the scores were (No) if the image quality criterion was not fulfilled (motion) and (Yes) if the image quality criterion was fulfilled (no motion). Inter-rater agreement was measured using the Gwet's agreement coefficient (AC1) with a 95% confidence interval. The accuracy of the EOS was assessed using sensitivity, specificity, and the area under curve (AUC). Variables between the EOS and DR were compared using McNemar's test. Both Gwet's AC1 and corresponding 95%CI were computed using irrCAC library in R4.41 software.</p><p><strong>Results: </strong>The mean (range) age of patients when they underwent DR and EOS imaging was 11.2 (4-18) and 11.8 (5-17) years, respectively. EOS demonstrated almost perfect inter-rater reliability in the resolution criterion (AC1 = 0.99, 95% CI = 0.98,1.00, p < 0.001), while DR showed similar reliability in the motion artefacts (AC1 = 0.96, 95% CI = 0.92.00, p < 0.001). Intra-rater reliability was high for most criteria. EOS images demonstrated high sensitivity for all criteria and variable specificity, which equalled or exceeded that of DR images. Significant differences were found in some criteria, such as the coverage of the area of interest (chi-squared = 37.39, p < 0.001) and hip level (chi-squared statistic = 31.44, p < 0.001).</p><p><strong>Conclusion: </strong>The high AC1, sensitivity, and AUCs highlight the diagnostic image quality of the EOS imaging system for mechanical axis images, making it a valuable modality in clinical settings. Therefore, the EOS imaging system can be used instead of DR in this context, offering advantages such as lower irradiation, shorter scanning time, and the capability for 3D reconstruction.</p>","PeriodicalId":21783,"journal":{"name":"Skeletal Radiology","volume":" ","pages":"2577-2589"},"PeriodicalIF":2.5,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13522051/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148252951","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Benign metastasizing leiomyoma: a rare case of osseous involvement.","authors":"Rebecca H Chun, Christin A Tiegs-Heiden, Geoffrey Halling, Daniel A Adamo, Thomas D Atwell","doi":"10.1007/s00256-026-05277-1","DOIUrl":"10.1007/s00256-026-05277-1","url":null,"abstract":"<p><p>Benign metastasizing leiomyoma is a rare entity characterized by the metastasis of histologically benign smooth muscle tumors that originate from uterine leiomyomas. Patients are typically women with a history of uterine leiomyomas for which they have undergone operative treatment. Benign metastasizing leiomyoma most frequently involves the lungs but has been described in other locations. Osseous involvement is rare, with less than 20 cases reported in literature, the majority occurring in the spine. The current case is that of a 67-year-old female with a history of breast carcinoma with multiple lytic osseous lesions involving the pelvis and scapula, initially suspected to represent metastatic disease. Following multi-modality imaging workup including radiographs, CT, MRI, and PET/CT, biopsy revealed benign metastasizing leiomyoma. This case illustrates the multimodality imaging appearance of this rare entity. Radiographs and CT typically show well-defined lytic lesion(s) without aggressive periosteal reaction. On MRI, lesions are often T1 and T2 hypointense, with diffuse enhancement following the administration of contrast. Lesions typically display no or mild FDG uptake on PET/CT, however, varying levels of FDG uptake have been reported in literature. Benign metastasizing leiomyoma with osseous involvement is a rare but clinically significant entity that requires a high index of suspicion in women with a history of uterine leiomyomas and new bone lesions. Radiologists should be aware of this entity so that they may include benign metastasizing leiomyoma in the differential diagnosis when appropriate.</p>","PeriodicalId":21783,"journal":{"name":"Skeletal Radiology","volume":" ","pages":"2695-2700"},"PeriodicalIF":2.5,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148176703","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Photon-counting detector computed tomography (PCD-CT) in multiple myeloma: a systematic review and trial sequential meta-analysis on image quality and radiation dose.","authors":"Victor Arthur Ohannesian, Bruna Colabono Taneja, Adham do Amaral E Castro, Frederico Celestino Miranda, Carlos Henrique Tudino, Flavio Duarte Silva, Atul Kumar Taneja","doi":"10.1007/s00256-026-05242-y","DOIUrl":"10.1007/s00256-026-05242-y","url":null,"abstract":"<p><strong>Objectives: </strong>To systematically review and perform a meta-analysis comparing the effects of PCD-CT versus EID-CT on image quality (sharpness) and radiation dose (CTDIvol) in patients with bone lesions due to multiple myeloma (MM).</p><p><strong>Methods: </strong>A comprehensive search of PubMed, Embase, Scopus, and Cochrane Central was conducted from inception to October 2025. Studies comparing PCD-CT and EID-CT in MM patients were included. Methodological quality was assessed using ROBINS-I, and the certainty of evidence was assessed using GRADE.</p><p><strong>Results: </strong>A total of 41 studies were identified that matched our search criteria. After duplicate removal and screening, five studies (n = 170 patients) were included in the systematic review, with four contributing to the meta-analysis. PCD-CT showed a significant pooled mean difference in image sharpness (mean difference, + 0.99 points; 95% CI, 0.62-1.37; p < 0.001). PCD-CT also demonstrated a reduction in radiation dose (mean difference, -4.95 milligrays; 95% CI, -8.39 to -1.50; p = 0.005). Trial sequential analysis (TSA) confirmed stability of pooled estimates, with conclusive evidence for image sharpness improvement, and suggestive yet incomplete evidence for radiation dose reduction.</p><p><strong>Conclusion: </strong>Compared to EID-CT in MM, PCD-CT significantly improves subjective image sharpness as supported by trial sequential analysis. Conventional meta-analysis suggested a reduction in radiation dose with PCD-CT; however, trial sequential analysis indicated that the cumulative evidence remains inconclusive. Further large-scale studies are suggested to confirm the magnitude of the radiation dose reduction benefit.</p>","PeriodicalId":21783,"journal":{"name":"Skeletal Radiology","volume":" ","pages":"2503-2515"},"PeriodicalIF":2.5,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148043681","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Preoperative deltoid CT attenuation predicts clinical outcomes and influences treatment selection in geriatric proximal humerus fractures.","authors":"Mehmet Ali Bozca, Sinan Diril, Yusuf Altuntas, Enver Ipek, Malik Çelik, Cemal Kural","doi":"10.1007/s00256-026-05295-z","DOIUrl":"10.1007/s00256-026-05295-z","url":null,"abstract":"<p><strong>Objective: </strong>To evaluate the clinical relevance of CT-derived deltoid muscle attenuation as an imaging-based indicator of muscle quality and its association with clinical outcomes in geriatric patients with proximal humerus fractures.</p><p><strong>Materials and methods: </strong>In this retrospective multicenter cohort study, 118 patients aged 65 years and older with three- or four-part proximal humerus fractures were included. Patients were treated either non-operatively or with reverse shoulder arthroplasty. Deltoid muscle attenuation and cross-sectional area were measured on preoperative CT images at a standardized axial level. Patients were stratified according to a median attenuation threshold. Clinical outcomes at one year were assessed using the Constant-Murley score, Disabilities of the Arm, Shoulder and Hand score, Visual Analog Scale, and subjective shoulder value.</p><p><strong>Results: </strong>No significant correlation was observed between deltoid muscle cross-sectional area and attenuation. Patients with higher attenuation demonstrated significantly better functional outcomes, including higher Constant-Murley and subjective shoulder value scores and lower Disabilities of the Arm, Shoulder and Hand and Visual Analog Scale scores. In this subgroup, reverse shoulder arthroplasty was associated with superior outcomes compared to non-operative treatment. In contrast, among patients with lower attenuation, no significant differences were observed in functional scores between treatment groups, although lower pain levels and higher subjective satisfaction were noted following surgical treatment.</p><p><strong>Conclusion: </strong>Preoperative deltoid muscle attenuation derived from CT is associated with clinical outcomes in geriatric proximal humerus fractures and may serve as a practical imaging-based parameter for patient stratification. These findings support the potential role of muscle quality assessment in preoperative evaluation and treatment decision-making.</p>","PeriodicalId":21783,"journal":{"name":"Skeletal Radiology","volume":" ","pages":"2485-2494"},"PeriodicalIF":2.5,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148412913","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Diagnostic accuracy of artificial intelligence for spinopelvic parameters in standing total spine X-ray and limitations after fusion surgery.","authors":"Tobias Winkler, Thilo Khakzad, Matthias Pumberger, Friederike Schömig, Matthias Becher, Torsten Diekhoff","doi":"10.1007/s00256-026-05289-x","DOIUrl":"10.1007/s00256-026-05289-x","url":null,"abstract":"<p><strong>Objective: </strong>To evaluate the accuracy of a commercial deep learning algorithm in measuring spinopelvic parameters on full spine radiographs.</p><p><strong>Materials and methods: </strong>This retrospective study analyzed total spine X-rays from a clinical cohort, assessing spinopelvic parameters (coronal Cobb angle, thoracic kyphosis, lumbar lordosis, pelvic incidence, sagittal vertical axis). Measurements were performed by clinical radiologists, a trained research reader serving as ground truth, and a commercial AI software. Inter-rater reliability was determined via intraclass correlation coefficients between an orthopedic surgeon and the trained reader (n = 50 images). Performance of both the AI and radiologists was statistically compared to the ground truth using mean absolute error, intraclass correlation coefficients, the Spearman correlation, and diagnostic accuracy metrics (sensitivity, specificity, predictive values).</p><p><strong>Results: </strong>Four hundred ninety-five radiographs were analyzed. Inter-rater reliability between human raters, assessed via intraclass correlation coefficients, was excellent (0.94-1). Agreement between the algorithm and ground truth, also assessed via intraclass correlation coefficients, was good to excellent (0.79-0.99), except for lumbar lordosis (0.68, moderate). Mean absolute errors were lowest for coronal Cobb angles (4,6°; 95% confidence interval = 3.6-5.5°) and highest for lumbar lordosis (8.1°; 95% confidence interval = 6.9-9.3°). The Spearman rank correlations ranged from 0.74 to 0.99, and sensitivity was moderate to excellent (72.6-94.8). These results closely matched those obtained when comparing radiologists to ground truth. In a subgroup of patients with prior spinal fusion surgery, the correlation between the algorithm predictions and the ground truth was reduced, and measurement deviations were higher compared to non-instrumented patients.</p><p><strong>Conclusion: </strong>The commercial AI software predicted most spinopelvic parameters with good reliability and accuracy, coinciding with radiologists in clinical practice, however, showed limitations in patients with spinal instrumentation.</p>","PeriodicalId":21783,"journal":{"name":"Skeletal Radiology","volume":" ","pages":"2627-2635"},"PeriodicalIF":2.5,"publicationDate":"2026-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13522076/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148382763","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}