OBJECTIVE: To determine the performance of T2* cartilage mapping in diagnosing and assessing disease activity in early axial spondyloarthritis (axSpA), and to investigate the interaction of cartilage damage with clinical characteristics, sacroiliitis MRI scorings, and diffusion metrics.
METHODS: This prospective study included 83 axSpA patients and 37 no-axSpA patients. Clinical characteristics, the Assessment of SpondyloArthritis International Society-defined active sacroiliitis on MRI, and T2* SIJs values were recorded. In axSpA, disease activity was evaluated using the ankylosing spondylitis disease activity score-C-reactive protein; active sacroiliitis was evaluated using Spondyloarthritis Research Consortium of Canada, intravoxel incoherent motion, and diffusion kurtosis imaging; chronic sacroiliitis was assessed using composite structural damage score (CSDS) and structural score fat. Mann-Whitney U-test, Kruskal-Wallis test with false discovery rate (FDR), ROC curve, and linear regression were used for statistical analysis.
RESULTS: AxSpA patients had significantly higher T2*SIJs values than no-axSpA patients. (22.86 ± 2.42 ms vs 20.36 ± 1.30 ms, p < 0.001). The combination of T2*SIJs values and active sacroiliitis on MRI had the highest AUC for identifying axSpA. T2*SIJs values were significantly different between the inactive and very high, moderate and very high, high and very high, as well as inactive and high disease activity groups (all pFDR < 0.05). Dk (β = 0.48) and CSDS (β = 0.48) were independently associated with T2*SIJs values.
CONCLUSIONS: T2* values may be a promising biomarker for diagnosing and differentiating disease activity in early axSpA. Both acute and chronic sacroiliitis influence cartilage properties.
CONCLUSIONS: Sacroiliac joint cartilage abnormalities can be quantified with T2* relaxation time and allow better characterization of early axSpA.
CONCLUSIONS: T2* mapping may have value in evaluating axSpA. The combination of T2* values and active sacroiliitis on MRI enhances diagnostic performance for axSpA. Abnormalities measured with T2* values correlate with disease activity, acute sacroiliitis, and degree of structural damage.

OBJECTIVE: This study aimed to evaluate the performance of a deep learning radiomics (DLR) model, which integrates multimodal MRI features and clinical information, in diagnosing sacroiliitis related to axial spondyloarthritis (axSpA).
METHODS: A total of 485 patients diagnosed with sacroiliitis related to axSpA (n = 288) or non-sacroiliitis (n = 197) by sacroiliac joint (SIJ) MRI between May 2018 and October 2022 were retrospectively included in this study. The patients were randomly divided into training (n = 388) and testing (n = 97) cohorts. Data were collected using three MRI scanners. We applied a convolutional neural network (CNN) called 3D U-Net for automated SIJ segmentation. Additionally, three CNNs (ResNet50, ResNet101, and DenseNet121) were used to diagnose axSpA-related sacroiliitis using a single modality. The prediction results of all the CNN models across different modalities were integrated using a stacking method based on different algorithms to construct ensemble models, and the optimal ensemble model was used as DLR signature. A combined model incorporating DLR signature with clinical factors was developed using multivariable logistic regression. The performance of the models was evaluated using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA).
RESULTS: Automated deep learning-based segmentation and manual delineation showed good correlation. ResNet50, as the optimal basic model, achieved an area under the curve (AUC) and accuracy of 0.839 and 0.804, respectively. The combined model yielded the highest performance in diagnosing axSpA-related sacroiliitis (AUC: 0.910; accuracy: 0.856) and outperformed the best ensemble model (AUC: 0.868; accuracy: 0.825) (all P < 0.05). Moreover, the DCA showed good clinical utility in the combined model.
CONCLUSIONS: We developed a diagnostic model for axSpA-related sacroiliitis by combining the DLR signature with clinical factors, which resulted in excellent diagnostic performance.

Objective.Sacroiliitis is an early pathological manifestation of ankylosing spondylitis (AS), and a positive sacroiliitis test on imaging may help clinical practitioners diagnose AS early. Deep learning based automatic diagnosis algorithms can deliver grading findings for sacroiliitis, however, it requires a large amount of data with precise labels to train the model and lacks grading features visualization. In this paper, we aimed to propose a radiomics and deep learning based deep feature visualization positive diagnosis algorithm for sacroiliitis on CT scans. Visualization of grading features can enhance clinical interpretability with visual grading features, which assist doctors in diagnosis and treatment more effectively.Approach.The region of interest (ROI) is identified by segmenting the sacroiliac joint (SIJ) 3D CT images using a combination of the U-net model and certain statistical approaches. Then, in addition to extracting spatial and frequency domain features from ROI according to the radiographic manifestations of sacroiliitis, the radiomics features have also been integrated into the proposed encoder module to obtain a powerful encoder and extract features effectively. Finally, a multi-task learning technique and five-class labels are utilized to help with performing positive tests to reduce discrepancies in the evaluation of several radiologists.Main results.On our private dataset, proposed methods have obtained an accuracy rate of 87.3%, which is 9.8% higher than the baseline and consistent with assessments made by qualified medical professionals.Significance.The results of the ablation experiment and interpreting analysis demonstrated that the proposed methods are applied in automatic CT scan sacroiliitis diagnosis due to their excellently interpretable and portable advantages.

BACKGROUND: The Spondyloarthritis Research Consortium of Canada (SPARCC) scoring system is a sacroiliitis grading system.
OBJECTIVE: To develop a deep learning-based pipeline for grading sacroiliitis using the SPARCC scoring system.
METHODS: Prospective.
METHODS: The study included 389 participants (42.2-year-old, 44.6% female, 317/35/37 for training/validation/testing). A pretrained algorithm was used to differentiate image with/without sacroiliitis.
UNASSIGNED: 3-T, short tau inversion recovery (STIR) sequence, fast spine echo.
RESULTS: The regions of interest as ground truth for models\' training were identified by a rheumatologist (HYC, 10-year-experience) and a radiologist (KHL, 6-year-experience) using the Assessment of Spondyloarthritis International Society definition of MRI sacroiliitis independently. Another radiologist (YYL, 4.5-year-experience) solved the discrepancies. The bone marrow edema (BME) and sacroiliac region models were for segmentation. Frangi-filter detected vessels used as intense reference. Deep learning pipeline scored using SPARCC scoring system evaluating presence and features of BMEs. A rheumatologist (SCWC, 6-year-experience) and a radiologist (VWHL, 14-year-experience) scored using the SPARCC scoring system once. The radiologist (YYL) scored twice with 5-day interval.
METHODS: Independent samples t-tests and Chi-squared tests were used. Interobserver and intraobserver reliability by intraclass correlation coefficient (ICC) and Pearson coefficient evaluated consistency between readers and the deep learning pipeline. We evaluated the performance using sensitivity, accuracy, positive predictive value, and Dice coefficient. A P-value <0.05 was considered statistically significant.
RESULTS: The ICC and the Pearson coefficient between the SPARCC scores from three readers and the deep learning pipeline were 0.83 and 0.86, respectively. The sensitivity in identifying BME and accuracy of identifying SI joints and blood vessels was 0.83, 0.90, and 0.88, respectively. The dice coefficients were 0.82 (sacrum) and 0.80 (ilium).
CONCLUSIONS: The high consistency with human readers indicated that deep learning pipeline may provide a SPARCC-informed deep learning approach for scoring of STIR images in spondyloarthritis.
METHODS: 1 TECHNICAL EFFICACY: Stage 2.

UNASSIGNED: To analyze individually and interactively critical risk factors, which are closely related to low bone mineral density (BMD) in patient with ankylosing spondylitis (AS).
UNASSIGNED: A total of 249 AS patients who visited China-Japan Friendship Hospital were included in this training set. Patients with questionnaire data, blood samples, X-rays, and BMD were collected. Logistic regression analysis was employed to identify key risk factors for low BMD in different sites, and predictive accuracy was improved by incorporating the selected significant risk factors into the baseline model, which was then validated using a validation set. The interaction between risk factors was analyzed, and predictive nomograms for low BMD in different sites were established.
UNASSIGNED: There were 113 patients with normal BMD, and 136 patients with low BMD. AS patients with hip involvement are more likely to have low BMD in the total hip, whereas those without hip involvement are more prone to low BMD in the lumbar spine. Chest expansion, mSASSS, radiographic average grade of the sacroiliac joint, and hip involvement were significantly associated with low BMD of the femoral neck and total hip. Syndesmophytes, hip involvement and higher radiographic average grade of the sacroiliac joint increases the risk of low BMD of the femoral neck and total hip in an additive manner. Finally, a prediction model was constructed to predict the risk of low BMD in total hip and femoral neck.
UNASSIGNED: This study identified hip involvement was strongly associated with low BMD of the total hip in AS patients. Furthermore, the risk of low BMD of the femoral neck and total hip was found to increase in an additive manner with the presence of syndesmophytes, hip involvement, and severe sacroiliitis. This finding may help rheumatologists to identify AS patients who are at a high risk of developing low BMD and prompt early intervention to prevent fractures.

Objective: To summarize the clinical, radiological characteristics, and prognosis of infectious sacroiliitis in children. Methods: A case-control study was conducted, including 12 cases of infectious sacroiliitis diagnosed in the Rheumatology and Immunology Department of the Children\'s Hospital affiliated with the Capital Institute of Pediatrics from June 2018 to June 2023. These cases comprised the case group. Concurrently, 28 cases of pediatric idiopathic arthritis involving the sacroiliac joint in the same department served as the control group. Basic patient information, clinical features, laboratory parameters, and clinical treatment outcomes for both groups were collected and analyzed. Independent sample t-tests and chi-squared tests were used for inter-group comparisons. Results: Among the 12 cases in the case group, there were 5 males and 7 females, with a disease duration of 0.8 (0.5, 1.2) months. Nine patients presented with fever, and 1 patient had limping gait. Human leukocyte antigen (HLA)-B27 positivity was observed in 1 case, and there was no family history of ankylosing spondylitis. In the control group of 28 cases, there were 19 males and 9 females, with a disease duration of 7.0 (3.0, 17.0) months. One patient (4%) had fever, and 14 cases (50%) exhibited limping gait. HLA-B27 positivity was found in 18 cases (64%), and 18 cases (64%) had a family history of ankylosing spondylitis. The case group had higher white blood cell count (WBC), neutrophil ratio, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) levels, as well as a higher proportion of unilateral involvement on magnetic resonance imaging and bone destruction on CT compared to the control group ((11.1±6.2)×109 vs. (7.3±2.3)×109/L, 0.64±0.10 vs. 0.55±0.12, 72 (34, 86) vs. 18 (5, 41) mm/1 h, 24.6 (10.1, 67.3) mg/L vs. 3.6 (0.8, 15.0) mg/L, 11/12 vs. 36% (10/28), 9/12 vs. 11% (3/28), t=2.90, 3.07, Z=-2.94, -3.28, χ2=10.55, 16.53, all P<0.05). Conclusions: Pediatric infectious sacroiliitis often presents as unilateral involvement with a short disease history. Elevated WBC, CRP, and ESR, as well as a high rate of bone destruction, are also common characteristics.
目的： 总结儿童感染性骶髂关节炎的临床、影像学特点及预后。 方法： 病例对照研究，纳入2018年6月至2023年6月在首都儿科研究所附属儿童医院风湿免疫科就诊的12例感染性骶髂关节炎为病例组，同期同科室28例幼年型特发性关节炎累及骶髂关节患儿为对照组，收集并分析两组患儿的基本资料、临床特征、实验室指标及临床治疗转归情况，组间比较采用独立样本t检验、χ2检验。 结果： 病例组12例中男5例，女7例，病程0.8（0.5，1.2）个月，发热9例，跛行1例，人类白细胞抗原B27（HLA-B27）阳性1例，无强直性脊柱炎家族史。对照组28例中男19例，女9例，病程7.0（3.0，17.0）个月，发热1例（4%），跛行14例（50%），HLA-B27阳性18例（64%），强直性脊柱炎家族史18例（64%）；病例组白细胞计数（WBC）、中性粒细胞比例、红细胞沉降率（ESR）、C反应蛋白（CRP）、磁共振成像单侧受累及CT骨质破坏比例均高于对照组［（11.1±6.2）×109比（7.3±2.3）×109/L、0.64±0.10比0.55±0.12、72（34，86）比18（5，41）mm/1 h、24.6（10.1，67.3）比3.6（0.8，15.0）mg/L、11/12比36%（10/28），9/12比11%（3/28），t=2.90、3.07，Z=-2.94、-3.28，χ2=10.55、16.53，均P<0.05］。 结论： 儿童感染性骶髂关节炎多表现为单侧受累，病史短，白细胞计数、CRP、ESR升高，骨质破坏比例高。.

OBJECTIVE: To determine the prevalence of sacroiliac joint variants in patients with axial spondyloarthritis (axSpA) using MRI-based synthetic CT images and to evaluate their relationships with the presence of bone marrow edema, as this may potentially complicate diagnosing active sacroiliitis on MRI in patients with suspected axSpA.
METHODS: 172 patients were retrospectively included. All patients underwent MRI because of clinical suspicion of sacroiliitis. The diagnosis of axSpA was made by a tertiary hospital rheumatologist. Two readers independently determined the presence of bone marrow edema and the presence of one or more of the nine known sacroiliac joint (SIJ) variants.
RESULTS: SIJ variants were common in axSpA patients (82.9%) and the non-SpA group (85.4%); there were no significant differences in prevalence. Bone marrow edema was frequently found in axSpA (86.8%) and non-SpA patients (34%). AxSpA patients with SIJ variants (except for accessory joint) demonstrated 4 to 10 times higher odds for bone marrow edema, however not statistically significant. The more variants were present in this group, the higher the chance of bone marrow edema. However, some multicollinearity cannot be excluded, since bone marrow edema is very frequent in the axSpA group by definition.
CONCLUSIONS: SIJ variants are common in axSpA and non-SpA patients. SIJ variants were associated with higher prevalence of bone marrow edema in axSpA patients, potentially due to altered biomechanics, except for accessory joint which may act as a stabilizer.

Ankylosing spondylitis (AS) is a chronic inflammatory disease that causes inflammatory low back pain and may even limit activity. The grading diagnosis of sacroiliitis on imaging plays a central role in diagnosing AS. However, the grading diagnosis of sacroiliitis on computed tomography (CT) images is viewer-dependent and may vary between radiologists and medical institutions. In this study, we aimed to develop a fully automatic method to segment sacroiliac joint (SIJ) and further grading diagnose sacroiliitis associated with AS on CT. We studied 435 CT examinations from patients with AS and control at two hospitals. No-new-UNet (nnU-Net) was used to segment the SIJ, and a 3D convolutional neural network (CNN) was used to grade sacroiliitis with a three-class method, using the grading results of three veteran musculoskeletal radiologists as the ground truth. We defined grades 0-I as class 0, grade II as class 1, and grades III-IV as class 2 according to modified New York criteria. nnU-Net segmentation of SIJ achieved Dice, Jaccard, and relative volume difference (RVD) coefficients of 0.915, 0.851, and 0.040 with the validation set, respectively, and 0.889, 0.812, and 0.098 with the test set, respectively. The areas under the curves (AUCs) of classes 0, 1, and 2 using the 3D CNN were 0.91, 0.80, and 0.96 with the validation set, respectively, and 0.94, 0.82, and 0.93 with the test set, respectively. 3D CNN was superior to the junior and senior radiologists in the grading of class 1 for the validation set and inferior to expert for the test set (P < 0.05). The fully automatic method constructed in this study based on a convolutional neural network could be used for SIJ segmentation and then accurately grading and diagnosis of sacroiliitis associated with AS on CT images, especially for class 0 and class 2. The method for class 1 was less effective but still more accurate than that of the senior radiologist.

OBJECTIVE: To evaluate the feasibility and diagnostic accuracy of a deep learning network for detection of structural lesions of sacroiliitis on multicentre pelvic CT scans.
METHODS: Pelvic CT scans of 145 patients (81 female, 121 Ghent University/24 Alberta University, 18-87 years old, mean 40 ± 13 years, 2005-2021) with a clinical suspicion of sacroiliitis were retrospectively included. After manual sacroiliac joint (SIJ) segmentation and structural lesion annotation, a U-Net for SIJ segmentation and two separate convolutional neural networks (CNN) for erosion and ankylosis detection were trained. In-training validation and tenfold validation testing (U-Net-n = 10 × 58; CNN-n = 10 × 29) on a test dataset were performed to assess performance on a slice-by-slice and patient level (dice coefficient/accuracy/sensitivity/specificity/positive and negative predictive value/ROC AUC). Patient-level optimisation was applied to increase the performance regarding predefined statistical metrics. Gradient-weighted class activation mapping (Grad-CAM++) heatmap explainability analysis highlighted image parts with statistically important regions for algorithmic decisions.
RESULTS: Regarding SIJ segmentation, a dice coefficient of 0.75 was obtained in the test dataset. For slice-by-slice structural lesion detection, a sensitivity/specificity/ROC AUC of 95%/89%/0.92 and 93%/91%/0.91 were obtained in the test dataset for erosion and ankylosis detection, respectively. For patient-level lesion detection after pipeline optimisation for predefined statistical metrics, a sensitivity/specificity of 95%/85% and 82%/97% were obtained for erosion and ankylosis detection, respectively. Grad-CAM++  explainability analysis highlighted cortical edges as focus for pipeline decisions.
CONCLUSIONS: An optimised deep learning pipeline, including an explainability analysis, detects structural lesions of sacroiliitis on pelvic CT scans with excellent statistical performance on a slice-by-slice and patient level.
CONCLUSIONS: An optimised deep learning pipeline, including a robust explainability analysis, detects structural lesions of sacroiliitis on pelvic CT scans with excellent statistical metrics on a slice-by-slice and patient level.
CONCLUSIONS: • Structural lesions of sacroiliitis can be detected automatically in pelvic CT scans. • Both automatic segmentation and disease detection yield excellent statistical outcome metrics. • The algorithm takes decisions based on cortical edges, rendering an explainable solution.

OBJECTIVE: To analyze the changes in the levels of bone metabolism markers related to sacroiliac joint (SIJ) inflammation in patients with axial spondyloarthritis (axSpA) after treatment with imrecoxib and celecoxib and evaluate their relationship with clinical efficacy.
METHODS: A total of 120 patients with axSpA with at least 2 magnetic resonance imaging (MRI) SIJ scans during a 12-week follow-up were enrolled. The levels of bone metabolism markers, including dickkopf-1(DKK-1), sclerostin, vascular endothelial growth factor (VEGF), bone morphogenetic protein-2 (BMP-2), osteoprotegerin (OPG), noggin, β-catenin, and RUNX2, were measured twice, and their association with disease activity and the Spondyloarthritis Research Consortium of Canada (SPARCC) score for SIJ were analyzed by univariate analysis of covariance.
RESULTS: A total of 116 patients completed the follow-up. The levels of sclerostin, OPG, noggin, DKK-1, and RUNX2 were increased, whereas those of VEGF and β-catenin were decreased. The levels of sclerostin and OPG were negatively correlated with disease duration. The levels of VEGF and β-catenin were significantly decreased (F = 7.866, P = 0.003; F = 4.106, P = 0.047) in patients with disease remission. A decrease in ESR was significantly correlated with decreased levels of Runx2 and SPARCC scores, with the levels of sclerostin being significantly elevated in the SPARCC-reduced group. There were no statistically significant differences between the imrecoxib and celecoxib groups (P> 0.05).
CONCLUSIONS: Imrecoxib and celecoxib affect SIJ inflammation, disease activity, and the course of disease by regulating bone metabolism and angiogenesis in axSpA. Key Points •After treatment with imrecoxib and celecoxib, the levels of sclerostin, OPG, noggin, DKK-1, and RUNX2 were increased, whereas those of VEGF and β-catenin were decreased, correlating with the course of disease, disease activity, and SIJ inflammation. • A decrease in ESR was significantly correlated with a decrease in the levels of RUNX2 and SIJ inflammation.. • The levels of sclerostin were more significantly elevated in SIJ inflammation remission group.. •Imrecoxib and celecoxib affect SIJ inflammation by regulating bone metabolism and angiogenesis in axSpA..