Reliability assessment of the OMERACT whole-body magnetic resonance imaging scoring system for juvenile idiopathic arthritis

Inter-reader reliability of a new scoring system for evaluating joint inflammation and enthesitis in whole body MRI (WBMRI) in juvenile idiopathic arthritis was tested. The scoring system grades 732 item-region combinations of bone marrow and soft tissue changes for commonly involved joints and entheseal sites. Five radiologists rated 17 WBMRI scans through an online rating platform. Item-wise reliability was calculated for 117 items with non-zero scores in >10 % of readings. Interquartile ranges of the five-reader Kappa reliability coefficients were 0.58-0.73 (range: 0.36-0.88) for the joints, 0.65-0.81 (range: 0.39-0.95) for the entheses, and 0.62-0.75 (range: 0.60-0.76) for chronic nonbacterial osteomyelitis-like lesions.

ESR Essentials: Imaging of sacroiliitis-practice recommendations by ESSR

Sacroiliitis is commonly seen in patients with axial spondyloarthritis, in whom timely diagnosis and treatment are crucial to prevent irreversible structural damage. Imaging has a prominent place in the diagnostic process and several new imaging techniques have been examined for this purpose. We present a summary of updated evidence-based practice recommendations for imaging of sacroiliitis. MRI remains the imaging modality of choice for patients with suspected sacroiliitis, using at least four sequences: coronal oblique T1-weighted and fluid-sensitive sequences, a perpendicular axial oblique sequence, and a sequence for optimal evaluation of the bone-cartilage interface. Both active inflammatory and structural lesions should be described in the report, indicating location and extent. Radiography and CT, especially low-dose CT, are reasonable alternatives when MRI is unavailable, as patients are often young. This is particularly true to evaluate structural lesions, at which CT excels. Dual-energy CT with virtual non-calcium images can be used to depict bone marrow edema. Knowledge of normal imaging features in children (e.g., flaring, blurring, or irregular appearance of the articular surface) is essential for interpreting sacroiliac joint MRI in children because these normal processes can simulate disease. CLINICAL RELEVANCE STATEMENT: Sacroiliitis is a potentially debilitating disease if not diagnosed and treated promptly, before structural damage to the sacroiliac joints occurs. Imaging has a prominent place in the diagnostic process. We present a summary of practice recommendations for imaging of sacroiliitis, including several new imaging techniques. KEY POINTS: • MRI is the modality of choice for suspected inflammatory sacroiliitis, including a joint-line-specific sequence for optimal evaluation of the bone-cartilage interface to improve detection of erosions. • Radiography and CT (especially low-dose CT) are reasonable alternatives when MRI is unavailable. • Knowledge of normal imaging features in children is mandatory for interpretation of MRI of pediatric sacroiliac joints.

Association of anatomical variants of the sacroiliac joint with bone marrow edema in patients with axial spondyloarthritis

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.

CONCLUSION

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.

Vertebral Body Reshaping after Fractures: An Important Index of Recovery in Glucocorticoid-Treated Children

PURPOSE

In this 6-year study we identified factors associated with spontaneous vertebral body reshaping in glucocorticoid (GC)-treated children with leukemia, rheumatic disorders, and nephrotic syndrome.

METHODS

Subjects were 79 children (mean age 7.4 years) who had vertebral fracture (VF) evaluation on lateral spine radiographs at least 1 year after VF detection. VF were graded using the modified Genant semiquantitative method and fracture burden for individuals was quantified using the spinal deformity index (SDI; sum of grades from T4 to L4).

RESULTS

Sixty-five children (82.3%) underwent complete vertebral body reshaping (median time from VF detection to complete reshaping 1.3 years by Cox proportional hazard modeling). Of 237 VF, the majority (83.1%) ultimately reshaped, with 87.2% reshaping in the thoracic region vs 70.7% in the lumbar region (P = .004). Cox models showed that (1) every g/m2 increase in GC exposure in the first year after VF detection was associated with a 19% decline in the probability of reshaping; (2) each unit increase in the SDI at the time of VF detection was associated with a 19% decline in the probability of reshaping [hazard ratio (HR) = 0.81; 95% confidence interval (CI) = 0.71, 0.92; P = .001]; (3) each additional VF present at the time of VF detection reduced reshaping by 25% (HR = 0.75; 95% CI = 0.62, 0.90; P = .002); and (4) each higher grade of VF severity decreased reshaping by 65% (HR = 0.35; 95% CI = 0.21, 0.57; P < .001).

CONCLUSION

After experiencing a VF, children with higher GC exposure, higher SDI, more severe fractures, or lumbar VF were at increased risk for persistent vertebral deformity.

MRI-based synthetic CT for assessment of the bony elements of the sacroiliac joints in children

OBJECTIVES

The purpose of this study is to assess the equivalency of MRI-based synthetic CT (sCT) to conventional CT for sacroiliac joint bony morphology assessment in children.

METHODS

A prospective study was performed. Children who had (PET-)CT-scan underwent additional MRI. sCT-CT image quality was analyzed by two readers subjectively overall, semi-quantitatively in terms of cortical delineation, joint facet defects, growth plate fusion, ossified nuclei, lumbosacral transitional anomaly, and bony bridges, and quantitatively for disc space height, spinal canal width, and sacral vertebrae width and height. Cohen's kappa and equivalence analyses with Bland-Altman plots were calculated for categorical and continuous measures respectively.

RESULTS

Ten patients were included (6 boys; aged 9-16 years; mean age 14 years). Overall sCT image quality was rated good. Semi-quantitative assessment of cortical delineation of sacroiliac joints, bony bridges, and joint facet defects on the right iliac and sacral sides showed perfect agreement. Correlation was good to excellent (kappa 0.615-1) for the presence of lumbosacral transitional anomaly, fusion of sacral growth plates, joint facet defect, and presence of ossified nuclei. sCT-CT measurements were statistically equivalent and within the equivalence margins (-1-1 mm) for intervertebral disc space height and spinal canal width. Intra- and inter-reader reliability was excellent for quantitative assessment (0.806 < ICC < 0.998). For categorical scoring, kappa ranged from substantial to excellent (0.615-1).

CONCLUSION

sCT appears to be visually equivalent to CT for the assessment of pediatric sacroiliac joints. sCT may aid in visualizing sacroiliac joints compared to conventional MRI, with the benefit that no ionizing radiation is used, especially important in children.

CRITICAL RELEVANCE STATEMENT

MRI-based synthetic CT, a new technique that generates CT-like images without ionizing radiation, appears to be visually equivalent to CT for assessment of normal pediatric sacroiliac joints and can potentially assess structural damage as it clearly depicts bony cortex.

KEY POINTS

• MRI-based sCT is a new image technique that can generate CT-like images. • We found that sCT performs similarly to CT in displaying bony structures of pediatric sacroiliac joints. • sCT has already been clinically validated in the sacroiliac joints in adults. • sCT can potentially assess structural damage from erosions or ankylosis as it clearly depicts bony cortex.

Risk Factors Associated with Incident Vertebral Fractures in Steroid-treated Males with Duchenne Muscular Dystrophy

PURPOSE

Prevention of fractures is an unmet need in glucocorticoid (GC)-treated Duchenne muscular dystrophy. This study explored factors associated with incident vertebral fractures (VFs) to inform future fracture prevention efforts.

METHODS

VFs were evaluated prospectively at study baseline and 12 months on lateral spine radiographs in participants aged 4 to 25 years with Duchenne muscular dystrophy. Clinical factors were analyzed for their association with the change in Spinal Deformity Index (sum of the Genant-defined VF grades from T4 to L4) between baseline and 12 months.

RESULTS

Thirty-eight males were evaluated (mean ± SD age at baseline 11.0 ± 3.6 years; mean ± SD GC duration at baseline 4.1 ± 3.1 years; 74% ambulatory). Nine of 38 participants (24%) had 17 incident VFs, of which 3/17 VFs (18%) were moderate/severe. Participants with 12-month incident VF had lower mean ± SD baseline lumbar spine areal bone mineral density Z-scores (-2.9 ± 1.0 vs -1.9 ± 1.1; P = .049) and lower total body less head areal bone mineral density Z-scores (-3.1 ± 1.2 vs -1.6 ± 1.7; P = .036). Multivariable linear regression showed that at least 1 VF at baseline (P < .001), a higher number of antecedent non-VF (P < .001), and greater bone age delay at baseline (P = .027) were significant predictors of an increase in the Spinal Deformity Index from baseline to 12 months.

CONCLUSION

The observation that ≥ 1 prevalent VF and/or non-VF were the strongest predictors of incident VFs at 12 months supports the need for prevention of first fractures in this high-risk setting. Bone age delay, a marker of GC exposure, may assist in the prioritization of patients in efforts to prevent first fractures.

Patellofemoral joint geometry and osteoarthritis features 3-10 years after knee injury compared with uninjured knees

In this cross-sectional study, we compared patellofemoral geometry in individuals with a youth-sport-related intra-articular knee injury to uninjured individuals, and the association between patellofemoral geometry and magnetic resonance imaging (MRI)-defined osteoarthritis (OA) features. In the Youth Prevention of Early OA (PrE-OA) cohort, we assessed 10 patellofemoral geometry measures in individuals 3-10 years following injury compared with uninjured individuals of similar age, sex, and sport, using mixed effects linear regression. We also dichotomized geometry to identify extreme (>1.96 standard deviations) features and assessed likelihood of having extreme values using Poisson regression. Finally, we evaluated the associations between patellofemoral geometry with MRI-defined OA features using restricted cubic spline regression. Mean patellofemoral geometry did not differ substantially between groups. However, compared with uninjured individuals, injured individuals were more likely to have extremely large sulcus angle (prevalence ratio [PR] 3.9 [95% confidence interval, CI: 2.3, 6.6]), and shallow lateral trochlear inclination (PR 4.3 (1.1, 17.9)) and trochlear depth (PR 5.3 (1.6, 17.4)). In both groups, high bisect offset (PR 1.7 [1.3, 2.1]) and sulcus angle (PR 4.0 [2.3, 7.0]) were associated with cartilage lesion, and most geometry measures were associated with at least one structural feature, especially cartilage lesions and osteophytes. We observed no interaction between geometry and injury. Certain patellofemoral geometry features are correlated with higher prevalence of structural lesions compared with injury alone, 3-10 years following knee injury. Hypotheses generated in this study, once further evaluated, could contribute to identifying higher-risk individuals who may benefit from targeted treatment aimed at preventing posttraumatic OA.

Neural network algorithm for detection of erosions and ankylosis on CT of the sacroiliac joints: multicentre development and validation of diagnostic accuracy

OBJECTIVES

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.

CLINICAL RELEVANCE STATEMENT

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.

KEY POINTS

• 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.

2D vs. 3D Ultrasound Diagnosis of Pediatric Supracondylar Fractures

Supracondylar fractures are common injuries in children. Diagnosis typically relies on radiography, which can involve long wait times in the ED, emits ionizing radiation, and can miss non-displaced fractures. Ultrasound (US) has the potential to be a safer, more convenient diagnostic tool, especially with new highly portable handheld 2D point of care US (POCUS). This study aimed to determine the reliability of 2D POCUS for the detection of supracondylar fractures and elbow joint effusions, to contrast the accuracy of 2D POCUS vs. 3DUS vs. radiographs, and to determine whether blinded image interpretation could produce similar results to non-blinded real-time imaging. Fifty-seven children were scanned with 2D POCUS and 3DUS on the affected elbow. US scans were then read by three blinded readers, and the results were compared to gold-standard radiographs. Compared to a gold standard of 30-day radiographic diagnosis, readers of 2D POCUS detected supracondylar fracture and effusion with sensitivities of 0.91 and 0.97, respectively, which were both higher than with 3DUS. Inter-rater reliability of fracture detection was moderate for 2D POCUS (k = 0.40) and 3DUS (k = 0.53). Consensus sensitivities, although high, were lower than reports from some non-blinded studies, indicating that clinical presentation serves as an important factor in detection rates. Our results from consensus US diagnosis support the validity of using 2D POCUS in children for supracondylar fracture and elbow effusion diagnosis.

The use of slice encoding for metal artifact correction (SEMAC) sequencing improves the diagnostic evaluation of graft integrity following anterior cruciate ligament reconstruction

OBJECTIVES

To determine whether magnetic resonance imaging (MRI) with metal artifact reduction sequencing is superior to conventional knee MRI in the evaluation of an injured anterior cruciate ligament (ACL) graft, where visualisation on conventional MRI can be limited by the metal artifact from fixation devices.

METHODS

Eighteen patients underwent conventional MRI sequence (proton density fat saturated [PDFS]) and two types of metal artifact reduction sequencing MRI (WARP, slice encoding for metal artifact correction (SEMAC); Siemens) following a secondary injury to their ACL reconstructed knee. Six raters with experience in knee MRI evaluation reviewed sagittal PDFS, WARP, and SEMAC sequences, providing semi-quantitative grades for visualisation and diagnostic confidence assessing the ACL, posterior cruciate ligament , menisci, tibial and femoral tunnel margins, and articular cartilage. Intra-class correlation coefficients for inter-rater reliability were evaluated. The 6-rater mean scores for the visualisation and diagnostic confidence derived from each sequence were compared using the Friedman test for multiple paired samples.

RESULTS

No statistically significant difference in the ACL visualisation among the sequences was found (p ​= ​0.193). Further, a subgroup analysis was performed in cases evaluated as "moderately blurry" or "indistinct ACL visualisation" on PDFS (58% of cases). SEMAC significantly improved diagnostic confidence in ACL visualisation (p ​= ​0.041) and ACL graft rupture (p ​= ​0.044) compared to PDFS. There was no statistically significant difference in the inter-observer reliability between sequences. The WARP sequence added 2.84 ​± ​0.69 ​min, while SEMAC added 2.95 ​± ​0.40 ​min to the standard knee MRI scan time.

CONCLUSION

use of the SEMAC metal reduction sequence significantly improved diagnostic accuracy and confidence in the detection of ACL graft rupture in cases where the ACL was moderately blurry or indistinct on the PDFS sequence. This sequence should be considered as an adjunct to conventional PDFS in cases where graft visualisation is limited by the metal artifact from fixation devices.

LEVEL OF EVIDENCE

III.

Self-supervised-RCNN for medical image segmentation with limited data annotation

Many successful methods developed for medical image analysis based on machine learning use supervised learning approaches, which often require large datasets annotated by experts to achieve high accuracy. However, medical data annotation is time-consuming and expensive, especially for segmentation tasks. To overcome the problem of learning with limited labeled medical image data, an alternative deep learning training strategy based on self-supervised pretraining on unlabeled imaging data is proposed in this work. For the pretraining, different distortions are arbitrarily applied to random areas of unlabeled images. Next, a Mask-RCNN architecture is trained to localize the distortion location and recover the original image pixels. This pretrained model is assumed to gain knowledge of the relevant texture in the images from the self-supervised pretraining on unlabeled imaging data. This provides a good basis for fine-tuning the model to segment the structure of interest using a limited amount of labeled training data. The effectiveness of the proposed method in different pretraining and fine-tuning scenarios was evaluated based on the Osteoarthritis Initiative dataset with the aim of segmenting effusions in MRI datasets of the knee. Applying the proposed self-supervised pretraining method improved the Dice score by up to 18% compared to training the models using only the limited annotated data. The proposed self-supervised learning approach can be applied to many other medical image analysis tasks including anomaly detection, segmentation, and classification.

Reductions in Bone Mineral Density Are Apparent Early in Children With Prevalent Osteonecrosis Lesions Following Leukemia Therapy

Osteonecrosis (ON) is a serious complication of childhood acute lymphoblastic leukemia. We determined the prevalence of osteonecrotic lesions in our patient population by a one-time multisite magnetic resonance imaging (MRI) more than 1 year following leukemia therapy. MRI findings were evaluated in relationship to clinical factors (including longitudinal changes in bone mineral density [BMD]). Eighty-six children enrolled in the Steroid Associated Osteoporosis in the Pediatric Population (STOPP) study were evaluated for ON at 3.1 ± 1.3 years following therapy. Thirty children had a total of 150 confirmed ON lesions (35%). Lumbar spine (LS) BMD Z-scores (mean ± SD) were low at diagnosis and similar between patients with and without ON (-1.09 ± 1.53 versus -1.27 ± 1.25, p = 0.549). LS BMD Z-scores declined from baseline to 12 months in children with ON (-0.31 ± 1.02) but not in those without (0.13 ± 0.82, p = 0.035); the hip BMD Z-scores from baseline to 24 months declined in both groups, but to a greater extent in those with ON (-1.77 ± 1.22) compared to those without (-1.03 ± 1.07, p = 0.045). At the time of the MRI, mean total hip and total body (TB) BMD Z-scores were lower in children with ON (hip -0.98 ± 0.95 versus -0.28 ± 1.06, p = 0.010; TB -1.36 ± 1.10 versus -0.48 ± 1.50, p = 0.018). Pain occurred in 11/30 (37%) with ON versus 20/56 (36%) without, p = 0.841. In multivariable models, older age at diagnosis (odds ratio [OR] 1.57; 95% confidence interval [CI], 1.15-2.13; p = 0.004), and hip BMD Z-score at MRI (OR 2.23; 95% CI, 1.02-4.87; p = 0.046) were independently associated with ON. Overall, one-third of children demonstrated ON after leukemia therapy. Those with ON had greater reductions in spine and hip BMD Z-scores in the first 1 and 2 years of therapy, respectively. Older age and lower hip BMD Z-scores at MRI were significantly associated with prevalent, off-therapy ON. These data assist in identifying children at risk of ON. © 2023 The Authors. Journal of Bone and Mineral Research published by Wiley Periodicals LLC on behalf of American Society for Bone and Mineral Research (ASBMR).

Comparative validation of the knee inflammation MRI scoring system and the MRI osteoarthritis knee score for semi-quantitative assessment of bone marrow lesions and synovitis-effusion in osteoarthritis: an international multi-reader exercise

Background

Bone marrow lesions (BMLs) and synovitis on magnetic resonance imaging (MRI) are associated with symptoms and predict degeneration of articular cartilage in osteoarthritis (OA). Validated methods for their semiquantitative assessment on MRI are available, but they all have similar scoring designs and questionable sensitivity to change. New scoring methods with completely different designs need to be developed and compared to existing methods.

Objectives

To compare the performance of new web-based versions of the Knee Inflammation MRI Scoring System (KIMRISS) with the MRI OA Knee Score (MOAKS) for quantification of BMLs and synovitis-effusion (S-E).

Design

Retrospective follow-up cohort.

Methods

We designed web-based overlays outlining regions in the knee that are scored for BML in MOAKS and KIMRISS. For KIMRISS, both BML and S-E are scored on consecutive sagittal slices. The performance of these methods was compared in an international reading exercise of 8 readers evaluating 60 pairs of scans conducted 1 year apart from cases recruited to the OA Initiative (OAI) cohort. Interobserver reliability for baseline status and baseline to 1 year change in BML and S-E was assessed by intra-class correlation coefficient (ICC) and smallest detectable change (SDC). Feasibility was assessed using the System Usability Scale (SUS).

Results

Mean change in BML and S-E was minimal over 1 year. Pre-specified targets for acceptable reliability (ICC ⩾ 0.80 and ⩾ 0.70 for status and change scores, respectively) were achieved more frequently for KIMRISS for both BML and synovitis. Mean (95% CI) ICC for change in BML was 0.88 (0.83-0.92) and 0.69 (0.60-0.78) for KIMRISS and MOAKS, respectively. KIMRISS mean SUS usability score was 85.7 and at the 95th centile of ranking for usability versus a score of 55.4 and 20th centile for MOAKS.

Conclusion

KIMRISS had superior performance metrics to MOAKS for quantification of BML and S-E. Both methods should be further compared in trials of new therapies for OA.

MRI in pediatric sacroiliitis, what radiologists should know

MRI is used for early detection of inflammation of sacroiliac joints as it shows active lesions of sacroiliitis long before radiographs show damage to the sacroiliac joints. Early diagnosis of arthritis allows early treatment of inflammation and can help delay disease progression and prevent irreversible damage. Also, early identification of axial involvement in juvenile spondyloarthropathy is crucial, as treatment options are different than for peripheral juvenile spondyloarthropathy. In general, standard sequences used in adults are also applied to children. However, interpreting MR images of pediatric sacroiliac joints is more challenging than in adults, because of normal physiological changes during skeletal maturation, which can simulate disease on MR images. Furthermore, classical definitions of sacroiliitis used in adults, for both active inflammatory and structural lesions, can be difficult to extrapolate to children. The development of reliable pediatric-specific definitions for sacroiliitis is still in active study. Understanding both normal and pathological signal changes in children is important to distinguish physiologic findings from disease and to make a correct diagnosis. In this review, the main imaging characteristics of sacroiliitis on MRI in children and its frequent pitfalls will be illustrated, while also citing some discussion points regarding the scan protocol.

Knee- and Overall Health-Related Quality of Life Following Anterior Cruciate Ligament Injury: A Cross-sectional Analysis of Australian and Canadian Cohorts

OBJECTIVE: To describe the knee- and overall health-related quality of life (QOL) 3 to 12 years after anterior cruciate ligament (ACL) tear, and to assess the association of clinical and structural features with QOL after ACL tear. DESIGN: Cross-sectional analysis of combined data from Australian (n = 76, 5.4 years postinjury) and Canadian (n = 50, 6.6 years postinjury) prospective cohort studies. METHODS: We conducted a secondary analysis of patient-reported outcomes and index knee magnetic resonance imaging (MRI) acquired in 126 patients (median 5.5 [range: 4-12] years postinjury), all treated with ACL reconstruction. Outcomes included knee (ACL Quality of Life questionnaire [ACL-QOL]) and overall health-related QOL (EQ-5D-3L). Explanatory variables were self-reported knee pain (Knee Injury and Osteoarthritis Outcome Score [KOOS-Pain subscale]) and function (KOOS-Sport subscale), and any knee cartilage lesion (MRI Osteoarthritis Knee Score). Generalized linear models were adjusted for clustering between sites. Covariates were age, sex, time since injury, injury type, subsequent knee injuries, and body mass index. RESULTS: The median [range] ACL-QOL score was 82 [24-100] and EQ-5D-3L was 1.0 [-0.2 to 1.0]. For every 10-point higher KOOS-Sport score, the ACL-QOL score increased by 3.7 points (95% confidence interval [CI]: 1.7, 5.7), whereas there was no evidence of an association with the EQ-5D-3L (0.00 points, 95% CI: -0.02, 0.02). There were no significant association between KOOS-Pain and ACL-QOL (4.9 points, 95% CI: -0.1, 9.9) or EQ-5D-3L (0.05 points, 95% CI: -0.01, 0.11), respectively. Cartilage lesions were not associated with ACL-QOL (-1.2, 95% CI: -5.1, 2.7) or EQ-5D-3L (0.01, 95% CI: -0.01, 0.04). CONCLUSION: Self-reported function was more relevant for knee-related QOL than knee pain or cartilage lesions after ACL tear. Self-reported function, pain, and knee structural changes were not associated with overall health-related QOL. J Orthop Sports Phys Ther 2023;53(7):1-12. Epub: 8 June 2023. doi:10.2519/jospt.2023.11838.

AI aided workflow for hip dysplasia screening using ultrasound in primary care clinics

Developmental dysplasia of the hip (DDH) is a common cause of premature osteoarthritis. This osteoarthritis can be prevented if DDH is detected by ultrasound and treated in infancy, but universal DDH screening is generally not cost-effective due to the need for experts to perform the scans. The purpose of our study was to evaluate the feasibility of having non-expert primary care clinic staff perform DDH ultrasound using handheld ultrasound with artificial intelligence (AI) decision support. We performed an implementation study evaluating the FDA-cleared MEDO-Hip AI app interpreting cine-sweep images obtained from handheld Philips Lumify probe to detect DDH. Initial scans were done by nurses or family physicians in 3 primary care clinics, trained by video, powerpoint slides and brief in-person. When the AI app recommended follow-up (FU), we first performed internal FU by a sonographer using the AI app; cases still considered abnormal by AI were referred to pediatric orthopedic clinic for assessment. We performed 369 scans in 306 infants. Internal FU rates were initially 40% for nurses and 20% for physicians, declining steeply to 14% after ~ 60 cases/site: 4% technical failure, 8% normal at sonographer FU using AI, and 2% confirmed DDH. Of 6 infants referred to pediatric orthopedic clinic, all were treated for DDH (100% specificity); 4 had no risk factors and may not have otherwise been identified. Real-time AI decision support and a simplified portable ultrasound protocol enabled lightly trained primary care clinic staff to perform hip dysplasia screening with FU and case detection rates similar to costly formal ultrasound screening, where the US scan is performed by a sonographer and interpreted by a radiologist/orthopedic surgeon. This highlights the potential utility of AI-supported portable ultrasound in primary care.

Determination of Relative Weightings for Sacroiliac Joint Pathologies in the OMERACT Juvenile Arthritis Magnetic Resonance Imaging Sacroiliac Joint Score

This study aims to determine the relative weights (point value) of items of the juvenile idiopathic arthritis magnetic resonance imaging-sacroiliac joint scoring system (JAMRIS-SIJ). An adaptive multicriteria decision analysis was performed using the 1000Minds web application to determine the relative weights of the items in the JAMRIS-SIJ inflammation and damage domains. Experts in imaging and rheumatology independently completed a conjoint analysis survey (CAS) to determine the point value of the measurement items of the JAMRIS-SIJ. Each CAS survey question asked the expert to compare two hypothetical patient profiles, which were otherwise similar but different at two items at a time, and to select which item showed a more severe stage of inflammation or osteochondral damage. In addition, experts ranked 14 JAMRIS-SIJ grade only or image + grade patient vignettes while blinded to the CAS-derived weights. The validity of the weighted JAMRIS-SIJ was tested by comparing the expert CAS-weighted score and the image + grade ranking method. Seventeen experts completed the CAS (11 radiologists and 6 rheumatologists). Considering the point value for inflammation domain items, osteitis (24.7%) and bone marrow edema (24.3%) had higher group-averaged percentage weights compared to inflammation in erosion cavity (16.9%), joint space enhancement (13.1%), joint space fluid (9.1%), capsulitis (7.3%), and enthesitis (4.6%). Similarly, concerning the damage domain, ankylosis (41.3%) and erosion (25.1%) showed higher group-averaged weights compared to backfill (13.9%), sclerosis (10.7%), and fat metaplasia lesion (9.1%). The Spearman correlation coefficients of the CAS-weighted vignette order and unweighted JAMRIS-SIJ grade only order vignettes for all experts were 0.79 for inflammation and 0.80 for damage. The correlations of image vignettes among imaging experts to CAS were 0.75 for inflammation and 0.90 for damage. The multicriteria decision analysis identified differences in relative weights among the JAMRIS-SIJ measurement items. The determination of the relative weights provided expert-driven score scaling and face validity for the JAMRIS-SIJ, enabling the future evaluation of its longitudinal construct validity.

Anatomical variation of the sacroiliac joints: an MRI study with synthetic CT images

BACKGROUND

Synthetic computed tomography (sCT) images are magnetic resonance imaging (MRI)-based images, generated using artificial intelligence. This study aimed to determine the prevalence of anatomical variants of sacroiliac joints (SIJ) on sCT images and the correlation with age, sex and body weight.

METHODS

MRI of the SIJ including sCT images of 215 patients clinically suspected for sacroiliitis were retrospectively analyzed. The presence of anatomical variants of the SIJ was assessed. Age, sex and body mass index at the time of the MRI were recorded.

RESULTS

SIJ variants were found in 82.8% (356/430) of the evaluated joints. The most frequent variants were iliosacral complex (27.7%), bipartite iliac bony plate (27.2%) and crescent iliac bony plate (27%). One new variant was identified, consisting of an accessory facet of the SIJ on the superior side. Overall, SIJ variants were slightly more frequent in women (85.8% vs. 77.8%), but iliosacral complex was significantly more frequent in men. Isolated synostosis was more prevalent with advancing age, in contrast to semicircular defect and unfused ossification center. The occurrence of iliosacral complex was associated with higher BMI, while crescent iliac bony plate occurred more in patients with lower BMI.

CONCLUSION

Over 80% of patients in this study, who were all suspected of sacroiliitis, had at least one SIJ variant. These variants may actually represent subtypes of the normal SIJ. sCT enables detection of very small or subtle findings including SIJ variants.

Risk factors associated with prevalent vertebral fractures in Duchenne muscular dystrophy

Patients with Duchenne muscular dystrophy (DMD) have a high fracture burden due to progressive myopathy and steroid-induced osteoporosis. This study in males with DMD showed that markers of systemic glucocorticoid exposure including shorter stature, greater bone age delay, and lower lumbar spine bone mineral density were associated with spine fragility.

INTRODUCTION

Fragility  fractures are frequent in DMD. The purpose of this study was to identify clinical factors associated with prevalent vertebral fractures (VF) in boys, teens/young adults with Duchenne muscular dystrophy (DMD).

METHODS

This was a cross-sectional study of males aged 4-25 years with DMD. VF were evaluated using the modified Genant semi-quantitative method on T4-L4 lateral spine radiographs. Areal bone mineral density (aBMD) was measured at the lumbar spine (LS) and used to estimate volumetric BMD (vBMD). Clinical factors were analyzed for their association with the Spinal Deformity Index (SDI, the sum of the Genant grades).

RESULTS

Sixty participants were enrolled (mean age 11.5 years, range 5.4-19.5). Nineteen participants (32%) had a total of 67 VF; 23/67 VF (34%) were moderate or severe. Participants with VF were shorter (mean height Z-score ± standard deviation: - 3.1 ± 1.4 vs. - 1.8 ± 1.4, p = 0.001), had longer glucocorticoid exposure (mean duration 6.0 ± 3.3 vs. 3.9 ± 3.3 years, p = 0.027), greater bone age (BA) delay (mean BA to chronological age difference - 3.2 ± 3.4 vs. - 1.3 ± 1.2 years, p = 0.035), and lower LSaBMD Z-scores (mean - 3.0 ± 1.0 vs. - 2.2 ± 1.2, p = 0.023). There was no difference in LSvBMD Z-scores. Multivariable Poisson regression showed that every 0.1 mg/kg/day increment in average glucocorticoid daily dose was associated with a 1.4-fold SDI increase (95% confidence interval: 1.1-1.7, p = 0.013). Greater BA delay (p < 0.001), higher weight Z-score (p = 0.004), decreased height Z-score (p = 0.025), and lower LSvBMD Z-score (p = 0.025) were also associated with SDI increase.

CONCLUSION

Readily measurable clinical variables were associated with prevalent VF in males with glucocorticoid-treated DMD. These variables may be useful to identify candidates for primary osteoporosis prevention after glucocorticoid initiation.

Blurring and Irregularity of the Subchondral Cortex in Pediatric Sacroiliac Joints on T1 Images: Incidence of Normal Findings That Can Mimic Erosions

OBJECTIVE

To determine prevalence of variations of subchondral bone appearance that may mimic erosions on T1-weighted magnetic resonance imaging (MRI) of pediatric sacroiliac (SI) joints according to age and sex.

METHODS

With ethics committee approval and informed consent, SI joint MRIs of 251 children (132 girls), mean age 12.4 years (range 6.1-18.0 years), were obtained in 2 cohorts: 127 children imaged for nonrheumatic reasons, and 124 children with low back pain but no features of sacroiliitis at initial clinical MRI review. MRIs were reviewed by 3 experienced radiologists, blinded from each other, for 3 features of the cortical black line representing the subchondral bone plate on T1-weighted MRI: visibility, blurring, and irregularity.

RESULTS

Based on agreement from 2 or more readers, the cortical black line was partially absent in 88.4% of the children, blurred in 34.7%, and irregular in 41.4%. All these features were most common on the iliac side of SI joints and at the first sacral vertebra level. Clearly visualized, sharply delineated SI joints with none of these features were seen in only 8.0% of children, or in 35.1% if we conservatively required agreement of all 3 readers to consider a feature present. There was no significant difference between sexes or cohorts; findings were similar across pediatric age groups.

CONCLUSION

Understanding the normal MRI appearance of the developing SI joint is necessary to distinguish physiologic findings from disease. At least two-thirds (65%) of normal pediatric SI joints showed at least 1 feature that is a component of the adult definition of SI joint erosions, risking overdiagnosis of sacroiliitis.

Temporomandibular joint segmentation in MRI images using deep learning

OBJECTIVES

Temporomandibular joint (TMJ) internal derangements (ID) represent the most prevalent temporomandibular joint disorder (TMD) in the population and its diagnosis typically relies on magnetic resonance imaging (MRI). TMJ articular discs in MRIs usually suffer from low resolution and contrast, and it is difficult to identify them. In this study, we applied two convolutional neural networks (CNN) to delineate mandibular condyle, articular eminence, and TMJ disc in MRI images.

METHODS

The models were trained on MRI images from 100 patients and validated on images from 40 patients using 2D slices and 3D volume as input, respectively. Data augmentation and five-fold cross-validation scheme were applied to further regularize the models. The accuracy of the models was then compared with four raters having different expertise in reading TMJ-MRI images to evaluate the performance of the models.

RESULTS

Both models performed well in segmenting the three anatomical structures. A Dice coefficient of about 0.7 for the articular disc, more than 0.9 for the mandibular condyle, and Hausdorff distance of about 2mm for the articular eminence were achieved in both models. The models reached near-expert performance for the segmentation of TMJ articular disc and performed close to the expert in the segmentation of mandibular condyle and articular eminence. They also surpassed non-experts in segmenting the three anatomical structures.

CONCLUSION

This study demonstrated that CNN-based segmentation models can be a reliable tool to assist clinicians identifying key anatomy on TMJ-MRIs. The approach also paves the way for automatic diagnosis of TMD.

CLINICAL SIGNIFICANCE

Accurately locating the articular disc is the hardest and most crucial step in the interpretation of TMJ-MRIs and consequently in the diagnosis of TMJ-ID. Automated software that assists in locating the articular disc and its surrounding structures would improve the reliability of TMJ-MRI interpretation, save time and assist in reader training. It will also serve as a foundation for additional automated analysis of pathology in TMJ structures to aid in TMD diagnosis.

Sickle cell bone disease and response to intravenous bisphosphonates in children

Children with sickle cell disease (SCD) have the potential for extensive and early-onset bone morbidity. This study reports on the diversity of bone morbidity seen in children with SCD followed at three tertiary centers. IV bisphosphonates were effective for bone pain analgesia and did not trigger sickle cell complications.

INTRODUCTION

To evaluate bone morbidity and the response to intravenous (IV) bisphosphonate therapy in children with SCD.

METHODS

We conducted a retrospective review of patient records from 2003 to 2019 at three Canadian pediatric tertiary care centers. Radiographs, magnetic resonance images, and computed tomography scans were reviewed for the presence of avascular necrosis (AVN), bone infarcts, and myositis. IV bisphosphonates were offered for bone pain management. Bone mineral density was assessed by dual-energy X-ray absorptiometry (DXA).

RESULTS

Forty-six children (20 girls, 43%) had bone morbidity at a mean age of 11.8 years (SD 3.9) including AVN of the femoral (17/46, 37%) and humeral (8/46, 17%) heads, H-shaped vertebral body deformities due to endplate infarcts (35/46, 76%), and non-vertebral body skeletal infarcts (15/46, 32%). Five children (5/26, 19%) had myositis overlying areas of AVN or bone infarcts visualized on magnetic resonance imaging. Twenty-three children (8/23 girls) received IV bisphosphonate therapy. They all reported significant or complete resolution of bone pain. There were no reports of sickle cell hemolytic crises, pain crises, or stroke attributed to IV bisphosphonate therapy.

CONCLUSION

Children with SCD have the potential for extensive and early-onset bone morbidity. In this series, IV bisphosphonates were effective for bone pain analgesia and did not trigger sickle cell complications.

ODP124 Sickle Cell Bone Disease and Response to Intravenous Bisphosphonates in Children

Purpose

To evaluate bone morbidity and the response to intravenous (IV) bisphosphonate therapy in children with Sickle Cell Disease (SCD).

Methods

We conducted a retrospective review of patient records from 2003 to 2019 at three Canadian pediatric tertiary care centers. Radiographs, magnetic resonance images, and computed tomography scans were reviewed for the presence of avascular necrosis (AVN), bone infarcts, and myositis. IV bisphosphonates were offered for bone pain management. Bone mineral density was assessed by dual-energy X-ray absorptiometry (DXA).

Results

Forty-six children (20 girls, 43%) had bone morbidity at a mean age of 11.8 years (SD 3.9) including AVN of the femoral (17/46, 37%) and humeral (8/46, 17%) heads, H-shaped vertebral body deformities due to endplate infarcts (35/46, 76%), and non-vertebral body skeletal infarcts (15/46, 32%). Five children (5/26, 19%) had myositis overlying areas of AVN or bone infarcts visualized on magnetic resonance imaging. Twenty-three children (8/23 girls) received IV bisphosphonate therapy. They all reported significant or complete resolution of bone pain. There were no reports of sickle cell hemolytic crises, pain crises or stroke attributed to IV bisphosphonate therapy. Bone mineral apparent density Z-score increased at the lumbar spine (N=14, mean change +0.6, SD 0.4) and total body less head (N=14, mean change +0.5, SD 0.4) after IV bisphosphonate therapy (mean 1.5 years, SD 0.8).

Conclusion

Children with SCD have the potential for extensive and early-onset bone morbidity. In this series, IV bisphosphonates were effective for bone pain analgesia and did not trigger sickle cell complications.

Presentation: No date and time listed

Unsupervised Multi-Latent Space RL Framework for Video Summarization in Ultrasound Imaging

The COVID-19 pandemic has highlighted the need for a tool to speed up triage in ultrasound scans and provide clinicians with fast access to relevant information. To this end, we propose a new unsupervised reinforcement learning (RL) framework with novel rewards to facilitate unsupervised learning by avoiding tedious and impractical manual labelling for summarizing ultrasound videos. The proposed framework is capable of delivering video summaries with classification labels and segmentations of key landmarks which enhances its utility as a triage tool in the emergency department (ED) and for use in telemedicine. Using an attention ensemble of encoders, the high dimensional image is projected into a low dimensional latent space in terms of: a) reduced distance with a normal or abnormal class (classifier encoder), b) following a topology of landmarks (segmentation encoder), and c) the distance or topology agnostic latent representation (autoencoders). The summarization network is implemented using a bi-directional long short term memory (Bi-LSTM) which utilizes the latent space representation from the encoder. Validation is performed on lung ultrasound (LUS), that typically represent potential use cases in telemedicine and ED triage acquired from different medical centers across geographies (India and Spain). The proposed approach trained and tested on 126 LUS videos showed high agreement with the ground truth with an average precision of over 80% and average F₁ score of well over 44 ±1.7 %. The approach resulted in an average reduction in storage space of 77% which can ease bandwidth and storage requirements in telemedicine.

Data-Driven Magnetic Resonance Imaging Definitions for Active and Structural Sacroiliac Joint Lesions in Juvenile Spondyloarthritis Typical of Axial Disease: A Cross-Sectional International Study

OBJECTIVE

We aimed to determine quantitative sacroiliac (SI) joint magnetic resonance imaging (MRI) cutoffs for active and structural lesions that will be incorporated as imaging domains in classification criteria of axial disease in juvenile spondyloarthritis (SpA).

METHODS

MRI scans from an international cross-section of juvenile SpA patients were reviewed by 6 musculoskeletal imaging experts blinded to clinical details. Raters globally assessed the presence/absence of lesions typical of axial SpA and performed SI joint quadrant- or joint-based scoring. Sensitivity and specificity of lesion cutoffs were calculated using a rater majority (≥4 of 6 raters) on a global assessment of the presence/absence of active or structural lesions typical of axial SpA with high confidence as the reference standard. Cutoffs were validated in an independent cohort.

RESULTS

Imaging from 243 subjects, 61% male, median age 14.9 years, had sequences available for detailed MRI scoring. Optimal cutoffs for defining lesions typical of axial disease in juvenile SpA were: 1) inflammatory lesion: bone marrow edema in ≥3 SI joint quadrants across all SI joint MRI slices (sensitivity 98.6%, specificity 96.5%); 2) structural lesions: erosion in ≥3 quadrants or sclerosis or fat lesion in ≥2 SI joint quadrants or backfill or ankylosis in ≥2 joint halves across all SI joint MRI slices (sensitivity 98.6%, specificity 95.5%). Sensitivity and specificity of the optimal cutoffs in the validation cohort were excellent.

CONCLUSION

We propose data-driven cutoffs for active inflammatory and structural lesions on MRI typical of axial disease in juvenile SpA that have high specificity and sensitivity using central imaging global assessment as the reference standard and excellent reliability.

Concurrent validity and reliability of a semi-automated approach to measuring the magnetic resonance imaging morphology of the knee joint in active youth

Post-traumatic knee osteoarthritis is attributed to alterations in joint morphology, alignment, and biomechanics triggered by injury. While magnetic resonance (MR) imaging-based measures of joint morphology and alignment are relevant to understanding osteoarthritis risk, time consuming manual data extraction and measurement limit the number of outcomes that can be considered and deter widespread use. This paper describes the development and evaluation of a semi-automated software for measuring tibiofemoral and patellofemoral joint architecture using MR images from youth with and without a previous sport-related knee injury. After prompting users to identify and select key anatomical landmarks, the software can calculate 37 (14 tibiofemoral, 23 patellofemoral) relevant geometric features (morphology and alignment) based on established methods. To assess validity and reliability, 11 common geometric features were calculated from the knee MR images (proton density and proton density fat saturation sequences; 1.5 T) of 76 individuals with a 3–10-year history of youth sport-related knee injury and 76 uninjured controls. Spearman’s or Pearson’s correlation coefficients (95% CI) and Bland-Altman plots were used to assess the concurrent validity of the semi-automated software (novice rater) versus expert manual measurements, while intra-class correlation coefficients (ICC2,1; 95%CI), standard error of measurement (95%CI), 95% minimal detectable change, and Bland-Altman plots were used to assess the inter-rater reliability of the semi-automated software (novice vs resident radiologist rater). Correlation coefficients ranged between 0.89 (0.84, 0.92; Lateral Trochlear Inclination) and 0.97 (0.96, 0.98; Patellar Tilt Angle). ICC estimates ranged between 0.79 (0.63, 0.88; Lateral Patellar Tilt Angle) and 0.98 (0.95, 0.99; Bisect Offset). Bland-Altman plots did not reveal systematic bias. These measurement properties estimates are equal, if not better than previously reported methods suggesting that this novel semi-automated software is an accurate, reliable, and efficient alternative method for measuring large numbers of geometric features of the tibiofemoral and patellofemoral joints from MR studies.

Interobserver Variability of Hip Dysplasia Indices on Sweep Ultrasound for Novices, Experts, and Artificial Intelligence

BACKGROUND

Ultrasound for developmental dysplasia of the hip (DDH) is challenging for nonexperts to perform and interpret. Recording "sweep" images allows more complete hip assessment, suitable for automation by artificial intelligence (AI), but reliability has not been established. We assessed agreement between readers of varying experience and a commercial AI algorithm, in DDH detection from infant hip ultrasound sweeps.

METHODS

We selected a full spectrum of poor-to-excellent quality images and normal to severe dysplasia, in 240 hips (120 single 2-dimensional images, 120 sweeps). For 12 readers (radiologists, sonographers, clinicians and researchers; 3 were DDH subspecialists), and a ultrasound-FDA-cleared AI software package (Medo Hip), we calculated interobserver reliability for alpha angle measurements by intraclass correlation coefficient (ICC2,1) and for DDH classification by Randolph Kappa.

RESULTS

Alpha angle reliability was high for AI versus subspecialists (ICC=0.87 for sweeps, 0.90 for single images). For DDH diagnosis from sweeps, agreement was high between subspecialists (kappa=0.72), and moderate for nonsubspecialists (0.54) and AI (0.47). Agreement was higher for single images (kappa=0.80, 0.66, 0.49). AI reliability deteriorated more than human readers for the poorest-quality images. The agreement of radiologists and clinicians with the accepted standard, while still high, was significantly poorer for sweeps than 2D images (P<0.05).

CONCLUSIONS

In a challenging exercise representing the wide spectrum of image quality and reader experience seen in real-world hip ultrasound, agreement on DDH diagnosis from easily obtained sweeps was only slightly lower than from single images, likely because of the additional step of selecting the best image. AI performed similarly to a nonsubspecialist human reader but was more affected by low-quality images.

Metaphyseal and posterior rib fractures in osteogenesis imperfecta: Case report and review of the literature

Purpose

Metaphyseal corner fractures and posterior rib fractures are thought to only occur in settings of inflicted injury. We describe a case of siblings who presented with metaphyseal corner fractures and multiple posterior rib fractures who were later found to carry FKBP10 mutations, a rare cause of Osteogenesis Imperfecta (OI) known as Bruck syndrome. This clinical presentation led to a literature review examining fracture types in OI and inflicted injury.

Cases

A 15-month-old male presented with multiple healing fractures of varying ages including posterior rib and metaphyseal corner fractures with no history of significant trauma. He had joint laxity, short stature and Wormian bones. His diagnosis of Bruck Syndrome led to investigations in his sibling at birth, which demonstrated the same fracture pattern including multiple posterior rib and metaphyseal corner fractures. They both had pathogenic compound heterozygous FKBP10 variants.

Literature review and results

We performed a literature review evaluating the fracture pattern in cases investigated for inflicted injury and found to have OI. Fourteen articles reported 78 children with OI initially diagnosed as inflicted injury. Of these children, 71 (91%) were diagnosed with milder forms of OI (Sillence type I and IV). Sixty-four children (81%) had clinical signs of OI including blue sclera, dentinogenesis imperfecta, short stature, joint laxity and limb bowing. Fifteen (19%) children had fractures of high specificity for inflicted injury including metaphyseal corner fractures and posterior rib fractures and 58 (74%) had fractures of moderate specificity for inflicted injury such as bilateral fractures and fractures of different ages.

Conclusion

Metaphyseal corner fractures and posterior rib fractures are highly associated with inflicted injury, but they have been reported in children with OI. Bruck syndrome, a rare and severe form of OI can present with metaphyseal and posterior rib fractures, including at birth. When features of OI are present in children with metaphyseal corner fractures and/or posterior rib fractures are present, genetic testing may be warranted.

•

Metaphyseal corner and posterior rib fractures are classically associated with inflicted injury.

•

However, they can occur in children with Osteogenesis Imperfecta (OI)

•

These fracture types can also occur in Bruck syndrome, a rare form of OI

Health-Related Outcomes 3-15 Years Following Ankle Sprain Injury in Youth Sport: What Does the Future Hold?

BACKGROUND

This study examined the association between youth sport-related ankle sprain injury and health-related outcomes, 3-15 years postinjury.

METHODS

A historical cohort study in which uninjured controls were cluster-matched with injured cases. The primary outcome was self-reported Foot and Ankle Outcome Score (FAOS). Secondary outcomes included measures of adiposity, validated questionnaires for physical activity, athletic identity, fear of pain, and tests of strength, balance, and function.

RESULTS

We recruited 86 participants (median age of 23 years; 77% female); 50 with a time-loss ankle sprain, median of 8 years postinjury, and 36 uninjured controls cluster-matched by sex and sport. Based on mixed effects multivariable regression models, previously injured participants demonstrated poorer outcomes than controls on all 5 FAOS subscales regardless of sex and time since injury, with the largest differences observed in symptoms (-20.9, 99% CI: -29.5 to -12.3) and ankle-related quality of life (-25.3, 99% CI: -34.7 to -15.9) subscales. Injured participants also had poorer unipedal dynamic balance (-1.9, 99% CI: 3.5 to -0.2) and greater fear of pain (7.2, 99% CI: 0.9-13.4) compared with controls. No statistically significant differences were found for other secondary outcomes.

CONCLUSION

At 3-15 years following time-loss ankle sprain injury in youth sport, previously injured participants had more pain and symptoms, poorer self-reported function, ankle-related quality of life, reduced sport participation, balance, and greater fear of pain than controls. This underlines the need to promote the primary prevention of ankle sprains and secondary prevention of potential health consequences, including posttraumatic osteoarthritis.

LEVEL OF EVIDENCE

Level III, historical cohort study.

Magnetic resonance imaging findings in the normal pediatric sacroiliac joint space that can simulate disease

BACKGROUND

Magnetic resonance imaging (MRI) features of active sacroiliac joint inflammation include joint space fluid and enhancement, but it is unclear to what extent these are present in normal children.

OBJECTIVE

To describe normal MRI appearances of pediatric sacroiliac joint spaces in boys and girls of varying ages.

MATERIALS AND METHODS

In this ethics-approved prospective study, 251 children (119 boys, 132 girls; mean age: 12.4 years, range: 6.1-18.0 years), had both oblique-coronal T1-weighted and short tau inversion recovery (STIR) sacroiliac joint MRI. Of these, 127 were imaged for other reasons and had asymptomatic sacroiliac joints ("normal cohort") while 124 had low back pain with no features of sacroiliitis on initial clinical MRI review ("low-back-pain cohort"). Post-gadolinium T1-weighted sequences were available in 16/127 normal and 124/124 low-back-pain subjects. Three experienced radiologists scored high signal in the sacroiliac joint space on STIR (score 0=absent; 1=high signal compared to normal bone marrow present anywhere in the joint but not as bright as fluid [compared to vessels, cerebrospinal fluid]; 2=definite fluid signal in part of the joint; 3=definite fluid signal, entire vertical height, majority of slices) and, when available, joint space post-contrast enhancement (0=no high signal/enhancement; 1=thin, symmetrical, mildly increased linear high signal present in the joint space; 2=focal, thick or intense enhancement). Associations between joint signal scores, age, gender and sacral apophyseal closure were analysed.

RESULTS

Increased signal on STIR (score 1-3) was present in 74.7% of pediatric sacroiliac joint spaces, as intense as fluid in 18.4%. There was no significant difference in proportion by gender, side or cohort, but girls showed peak signal earlier than boys (10 years old vs. 12 years old, respectively). On post-gadolinium T1-weighted sequences, a thin rim of increased signal was nearly universally seen in sacroiliac joint spaces without focal, intense or thick post-contrast enhancement.

CONCLUSION

Sacroiliac joint spaces of most children demonstrate mildly increased signal on STIR, compared to normal bone marrow, and thin rim-like enhancement on post-contrast T1 images, likely related to cartilage. These findings should not be confused with sacroiliitis.

Zoledronic Acid vs Placebo in Pediatric Glucocorticoid-induced Osteoporosis: A Randomized, Double-blind, Phase 3 Trial

CONTEXT

Glucocorticoids (GCs) prescribed for chronic pediatric illnesses are associated with osteoporotic fractures.

OBJECTIVE

This study aims to determine the efficacy and safety of intravenous (IV) zoledronic acid (ZA) compared with placebo to treat pediatric GC-induced osteoporosis (GIO).

METHODS

Children aged 5 to 17 years with GIO were enrolled in this multinational, randomized, double-blind, placebo-controlled phase 3 trial (ClinicalTrials.gov NCT00799266). Eligible children were randomly assigned 1:1 to 6 monthly IV ZA 0.05 mg/kg or IV placebo. The primary end point was the change in lumbar spine bone mineral density z score (LSBMDZ) from baseline to month 12. Incident fractures and safety were assessed.

RESULTS

Thirty-four children were enrolled (mean age 12.6 ± 3.4 years [18 on ZA, 16 on placebo]), all with low-trauma vertebral fractures (VFs). LSBMDZ increased from -2.13 ± 0.79 to -1.49 ± 1.05 on ZA, compared with -2.38 ± 0.90 to -2.27 ± 1.03 on placebo (least squares means difference 0.41 [95% CI, 0.02-0.81; P = .04]); when corrected for height z score, the least squares means difference in LBMDZ was 0.75 [95% CI, 0.27-1.22; P = .004]. Two children on placebo had new low-trauma VF vs none on ZA. Adverse events (AEs) were reported in 15 of 18 children (83%) on ZA, and in 12 of 16 (75%) on placebo, most frequently within 10 days after the first infusion. There were no deaths or treatment discontinuations due to treatment-emergent AEs.

CONCLUSION

LSBMDZ increased significantly on ZA compared with placebo over 1 year in children with GIO. Most AEs occurred after the first infusion.

Osteoporotic Fractures and Vertebral Body Reshaping in Children With Glucocorticoid-treated Rheumatic Disorders

CONTEXT

Osteoporotic fractures are an important cause of morbidity in children with glucocorticoid-treated rheumatic disorders.

OBJECTIVE

This work aims to evaluate the incidence and predictors of osteoporotic fractures and potential for recovery over six years following glucocorticoid (GC) initiation in children with rheumatic disorders.

METHODS

Children with GC-treated rheumatic disorders were evaluated through a prospective inception cohort study led by the Canadian STeroid-induced Osteoporosis in the Pediatric Population (STOPP) Consortium. Clinical outcomes included lumbar spine bone mineral density (LS BMD), vertebral fractures (VF), non-VF, and vertebral body reshaping.

RESULTS

A total of 136 children with GC-treated rheumatic disorders were enrolled (mean age 9.9 years, SD 4.4). The 6-year cumulative fracture incidence was 16.3% for VF, and 10.1% for non-VF. GC exposure was highest in the first 6 months, and 24 of 38 VF (63%) occurred in the first 2 years. Following VF, 16 of 19 children (84%) had complete vertebral body reshaping. Increases in disease activity and body mass index z scores in the first year and declines in LS BMD z scores in the first 6 months predicted incident VF over the 6 years, while higher average daily GC doses predicted both incident VF and non-VF. LS BMD z scores were lowest at 6 months (mean -0.9, SD 1.2) and remained low by 6 years even when adjusted for height z scores (-0.6, SD 0.9).

CONCLUSION

VF occurred early and were more common than non-VF in children with GC-treated rheumatic disorders. Eighty-four percent of children with VF underwent complete vertebral body reshaping, whereas vertebral deformity persisted in the remainder of children. On average, LS BMD z scores remained low at 6 years, consistent with incomplete recovery.

Automatic Assessment Of Hip Effusion From MRI

Joint effusion is a hallmark of osteoarthritis (OA) associated with stiffness, and may relate to pain, disability, and long-term outcomes. However, it is difficult to quantify accurately. We propose a new Deep Learning (DL) approach for automatic effusion assessment from Magnetic Resonance Imaging (MRI) using volumetric quantification measures (VQM). We developed a new multiplane ensemble convolutional neural network (CNN) approach for 1) localizing bony anatomy and 2) detecting effusion regions. CNNs were trained on femoral head and effusion regions manually segmented from 3856 images (63 patients). Upon validation on a non-overlapping set of 2040 images (34 patients) DL showed high agreement with ground-truth in terms of Dice score (0.85), sensitivity (0.86) and precision (0.83). Agreement of VQM per-patient was high for DL vs experts in term of Intraclass correlation coefficient (ICC)= 0.88[0.80,0.93]. We expect this technique to reduce inter-observer variability in effusion assessment, reducing expert time and potentially improving the quality of OA care.Clinical Relevance- Our technique for automatic assessment of hip MRI can be used for volumetric measurement of effusion. We expect this to reduce variability in OA biomarker assessment and provide more reliable indicators for disease progression.

Integrating User-Input into Deep Convolutional Neural Networks for Thyroid Nodule Segmentation

Delineation of thyroid nodule boundaries is necessary for cancer risk assessment and accurate categorization of nodules. Clinicians often use manual or bounding-box approach for nodule assessment which leads to subjective results. Consequently, agreement in thyroid nodule categorization is poor even among experts. Computer-aided diagnosis systems could reduce this variability by minimizing the extent of user interaction and by providing precise nodule segmentations. In this study, we present a novel approach for effective thyroid nodule segmentation and tracking using a single user click on the region of interest. When a user clicks on an ultrasound sweep, our proposed model can predict nodule segmentation over the entire sequence of frames. Quantitative evaluations show that the proposed method out-performs the bounding box approach in terms of the dice score on a large dataset of 372 ultrasound images. The proposed approach saves expert time and reduces the potential variability in thyroid nodule assessment. The proposed one-click approach can save clinicians time required for annotating thyroid nodules within ultrasound images/sweeps. With minimal user interaction we would be able to identify the nodule boundary which can further be used for volumetric measurement and characterization of the nodule. This approach can also be extended for fast labeling of large thyroid imaging datasets suitable for training machine-learning based algorithms.

Pediatric Imaging of the Elbow: A Pictorial Review

The elbow is a complex joint, subject to a wide range of traumatic, inflammatory, metabolic and neoplastic insults. The pediatric elbow has several diagnostic pitfalls due to the normal developmental changes in children. Knowledge of these normal variants is essential for both diagnosis and management of their elbow injuries. Radiography remains the first imaging modality of choice. Magnetic resonance imaging is excellent in evaluating lesions within the bone and soft tissues. In this pictorial essay, we provide insights into pediatric elbow imaging, show a range of entities specific to the pediatric elbow, and discuss diagnostic pitfalls that result from normal elbow growth in children.

Narrative Review on the Role of Imaging in DDH

BACKGROUND

Developmental dysplasia of hip (DDH) represents a spectrum from acetabular dysplasia to fixed dislocation, giving disability through premature osteoarthritis. Most DDH cases continue to present without any known risk factors such as breech presentation, female sex, and family history. Incidence and population-based outcomes of DDH are difficult to reliably establish due to many DDH definitions and classifications using different types of examinations.

PURPOSE

This review takes a historical perspective on the role of imaging in DDH.

METHODS

Pelvic radiographs (X-Ray) were amongst the first medical images identifying DDH, but these have a limited role in infancy due to absent ossification. In the 1980s, ultrasound led to a large expansion in infant DDH screening. Unfortunately, even for well-trained users, DDH indices on ultrasound generally lack reproducibility, and have led to overdiagnosis of mild DDH. CT and MRI more thoroughly evaluate the 3D hip deformity in DDH, but are costly, less available and involve radiation dose and/or anaesthesia.

RESULTS

Recently 3D ultrasound has been used to characterize the 3D deformity of DDH more fully, with improved inter-observer reliability, particularly amongst novice users. 3D ultrasound is also well suited to automated image analysis, but high-resolution 3D probes are costly and not widely available.

CONCLUSION

Combining the latest handheld portable ultrasound probes and artificial intelligence analysis could lead to an inexpensive tool permitting practical mass population screening for DDH. Overall, our understanding of DDH is heavily influenced by the imaging tools used to visualize it and changing quickly with modern technology.

Assessing the Reliability of the OMERACT Juvenile Idiopathic Arthritis Magnetic Resonance Scoring System for Temporomandibular Joints (JAMRIS-TMJ)

Contrast-enhanced magnetic resonance imaging (MRI) remains the most comprehensive modality to assess juvenile idiopathic arthritis (JIA)-related inflammation and osteochondral damage in the temporomandibular joints (TMJ). This study tested the reliability of a new JIA MRI scoring system for TMJ (JAMRIS-TMJ) and the impact of variations in calibration and reader specialty. Thirty-one MRI exams of bilateral TMJs were scored independently using the JAMRIS-TMJ by 20 readers consisting of radiologists and non-radiologist clinicians in three reading groups, with or without a calibrating atlas and/or tutorial. The inter-reader reliability in the multidisciplinary cohort assessed by the generalizability coefficient was 0.61–0.67 for the inflammatory and 0.66–0.74 for the damage domain. The atlas and tutorial did not improve agreement within radiologists, but improved the agreement between radiologist and non-radiologist groups. Agreements between different calibration levels were 0.02 to 0.08 lower by the generalizability coefficient compared to agreement within calibration levels; agreement between specialty groups was 0.04 to 0.10 lower than within specialty groups. Averaging two radiologists raised the reliability above 0.8 for both domains. Therefore, the reliability of JAMRIS-TMJ was moderate-to-good depending on the presence of specialty and calibration differences. The atlas and tutorial are necessary to improve reliability when the reader cohort consists of multiple specialties.

Variability in Interpretation of Magnetic Resonance Imaging of the Pediatric Sacroiliac Joint

OBJECTIVE

Magnetic resonance imaging (MRI) is pivotal in the assessment of early sacroiliitis in children. We aimed to evaluate the agreement between local radiology reports and central imaging reviewers for active inflammation and structural damage at the sacroiliac (SI) joints.

METHODS

Eight hospitals each contributed up to 20 cases of consecutively imaged children and adolescents with juvenile idiopathic arthritis and suspected sacroiliitis. Studies were independently reviewed by 3 experienced musculoskeletal pediatric radiologists. Local assessments of global impression and lesions were coded from the local radiology reports by 2 study team members. Test properties of local reports were calculated using the central imaging team's majority as the reference standard.

RESULTS

For 120 evaluable subjects, the median age was 14 years, half of the cases were male, and median disease duration at the time of imaging was 0.8 years (interquartile range 0-2). Sensitivity of local reports for inflammation was high, 93.5% (95% confidence interval [95% CI] 78.6-99.2), and specificity was moderate, 69.7% (95% CI 59.0-79.0), but positive predictive value (PPV) was low, 51.8% (95% CI 38.0-65.3). Twenty-seven cases (23%) had active inflammation reported locally but rated normal at the central reading, 19 (70%) with subsequent medication changes. The sensitivity of local reports detecting structural damage was low, 45.7% (95% CI 28.8-63.4), and specificity was high, 88.2% (95% CI 79.4-94.2); PPV was low, 61.5% (95% CI 40.6-79.8).

CONCLUSION

Substantial variation exists in the interpretation of inflammatory and structural lesions at the SI joints in children. To reliably identify pathology, additional training in the MRI appearance of the maturing SI joint is greatly needed.

Artificial Intelligence to Automatically Assess Scan Quality in Hip Ultrasound

PURPOSE

Since it is fast, inexpensive and increasingly portable, ultrasound can be used for early detection of Developmental Dysplasia of the Hip (DDH) in infants at point-of-care. However, accurate interpretation\is highly dependent on scan quality. Poor-quality images lead to misdiagnosis, but inexperienced users may not even recognize the deficiencies in the images. Currently, users assess scan quality subjectively, based on image landmarks which are prone to human errors. Instead, we propose using Artificial Intelligence (AI) to automatically assess scan quality.

METHODS

We trained separate Convolutional Neural Network (CNN) models to detect presence of each of four commonly used ultrasound landmarks in each hip image: straight horizontal iliac wing, labrum, os ischium and midportion of the femoral head. We used 100 3D ultrasound (3DUS) images for training and validated the technique on a set of 107 3DUS images also scored for landmarks by three non-expert readers and one expert radiologist.

RESULTS

We got AI ≥ 85% accuracy for all four landmarks (ilium = 0.89, labrum = 0.94, os ischium = 0.85, femoral head = 0.98) as a binary classifier between adequate and inadequate scan quality. Our technique also showed excellent agreement with manual assessment in terms of Intraclass Correlation Coefficient (ICC) and Cohen's kappa coefficient (K) for ilium (ICC = 0.81, K = 0.56), os ischium (ICC = 0.89, K = 0.63) and femoral head (ICC = 0.83, K = 0.66), and moderate to good agreement for labrum (ICC = 0.65, K = 0.33).

CONCLUSION

This new technique could ensure high scan quality and facilitate more widespread use of ultrasound in population screening of DDH.

Atlas of MRI findings of sacroiliitis in pediatric sacroiliac joints to accompany the updated preliminary OMERACT pediatric JAMRIS (Juvenile Idiopathic Arthritis MRI Score) scoring system: Part I: Active lesions

Magnetic resonance imaging (MRI) is an increasingly important tool for identifying involvement of the sacroiliac joints (SIJ) in juvenile idiopathic arthritis (JIA). The key feature for diagnosing active sacroiliitis is bone marrow edema (BME), but other features of active arthritis such as joint space inflammation, inflammation in an erosion cavity, capsulitis and enthesitis can be seen as well. Structural changes may also be seen. Systematic MRI assessment of inflammation and structural damage may aid in monitoring the disease course, choice of therapeutics and evaluating treatment response. In this pictorial essay, we illustrate normal MRI findings and growth-related changes of the SIJ in the pediatric population, as well as the different MRI features of SIJ inflammation. This atlas demonstrates fundamental MRI disease features of active inflammation in a format that can serve as a reference for assessing SIJ arthritis according to the updated preliminary JAMRIS (Juvenile Idiopathic Arthritis MRI Score) scoring system proposed by the MRI in JIA working group of Outcome Measures in Rheumatology and Clinical Trials (OMERACT). The atlas is intended to be read in conjunction with its companion Part 2, Structural Lesions.

The Accuracy of Incident Vertebral Fracture Detection in Children Using Targeted Case-Finding Approaches

Vertebral fractures are clinically important sequelae of a wide array of pediatric diseases. In this study, we examined the accuracy of case-finding strategies for detecting incident vertebral fractures (IVF) over 2 years in glucocorticoid-treated children (n = 343) with leukemia, rheumatic disorders, or nephrotic syndrome. Two clinical situations were addressed: the prevalent vertebral fracture (PVF) scenario (when baseline PVF status was known), which assessed the utility of PVF and low lumbar spine bone mineral density (LS BMD; Z-score <-1.4), and the non-PVF scenario (when PVF status was unknown), which evaluated low LS BMD and back pain. LS BMD was measured by dual-energy X-ray absorptiometry, vertebral fractures were quantified on spine radiographs using the modified Genant semiquantitative method, and back pain was assessed by patient report. Forty-four patients (12.8%) had IVF. In the PVF scenario, both low LS BMD and PVF were significant predictors of IVF. Using PVF to determine which patients should have radiographs, 11% would undergo radiography (95% confidence interval [CI] 8-15) with 46% of IVF (95% CI 30-61) detected. Sensitivity would be higher with a strategy of PVF or low LS BMD at baseline (73%; 95% CI 57-85) but would require radiographs in 37% of children (95% CI 32-42). In the non-PVF scenario, the strategy of low LS BMD and back pain produced the highest specificity of any non-PVF model at 87% (95% CI 83-91), the greatest overall accuracy at 82% (95% CI 78-86), and the lowest radiography rate at 17% (95% CI 14-22). Low LS BMD or back pain in the non-PVF scenario produced the highest sensitivity at 82% (95% CI 67-92), but required radiographs in 65% (95% CI 60-70). These results provide guidance for targeting spine radiography in children at risk for IVF. © 2021 American Society for Bone and Mineral Research (ASBMR).

Joint and entheseal inflammation in the knee region in spondyloarthritis - reliability and responsiveness of two OMERACT whole-body MRI scores

OBJECTIVE

To perform region-based development of whole-body MRI through validation of knee region scoring systems in spondyloarthritis (SpA).

METHODS

Assessment of knee inflammatory pathologies using 2 systems, OMERACT MRI Whole-body score for Inflammation in Peripheral joints and Entheses (MRI-WIPE) and Knee Inflammation MRI Scoring System (KIMRISS), in 4 iterative multi-reader exercises.

RESULTS

In the final exercise, reliability was mostly good for readers with highest agreement in previous exercise. Median pairwise single-measure ICCs for osteitis and synovitis/effusion status/change were 0.71/0.48 (WIPE-osteitis), 0.48/0.77 (WIPE-synovitis/effusion), 0.59/0.91 (KIMRISS-osteitis) and 0.92/0.97 (KIMRISS-synovitis/effusion). SRMs were 0.74 (WIPE-synovitis/effusion) and 0.78 (KIMRISS-synovitis/effusion).

CONCLUSION

MRI-WIPE and KIMRISS may both be useful in SpA whole-body evaluation studies.

The OMERACT Knee Inflammation MRI Scoring System: Validation of quantitative methodologies and tri-compartmental overlays in osteoarthritis

OBJECTIVE

To validate a revised version of the KIMRISS method for quantification of BML and synovitis-effusion in the knee by comparison with an established method, MOAKS.

METHODS

Novel calibration tools were developed for both methods. We compared reliability for status and change scores of BML and synovitis-effusion on baseline and one-year MRI scans.

RESULTS

Significant increase in both BML and synovitis-effusion was evident using KIMRISS but only for synovitis-effusion using MOAKS. Pre-specified targets for acceptable reliability (≥0.80 and ≥0.70 for status and change scores, respectively) were achieved more frequently for KIMRISS for both BML and synovitis.

CONCLUSION

Per OFISA criteria, KIMRISS should progress to assessment of discrimination.

Arthritis and enthesitis in the hip and pelvis region in spondyloarthritis - OMERACT validation of two whole-body MRI methods

OBJECTIVE

To validate reliability, correlation and responsiveness of two whole-body MRI scores for the hip/pelvis region in spondyloarthritis.

METHODS

Assessment of hip/pelvis inflammation in 4 multi-reader exercises using the OMERACT MRI Whole-body score for Inflammation in Peripheral joints and Entheses (MRI-WIPE) and Hip Inflammation Magnetic Resonance Imaging Scoring System (HIMRISS).

RESULTS

In exercises 3-4 (11/20 cases, respectively; 9 readers) reliability was mostly good for the 3 best calibrated readers. Median pairwise single-measure ICC for status were 0.58-0.65 (WIPE-osteitis), 0.10-0.88 (HIMRISS-osteitis) and for status/change 0.38-0.72/0.52-0.60 (WIPE-synovitis/effusion) and 0.68-0.89/0.78-0.85 (HIMRISS-synovitis/effusion). SRM was 1.23 for WIPE-osteitis, while lower for WIPE-synovitis/effusion and HIMRISS.

CONCLUSION

MRI-WIPE and HIMRISS may after further validation be useful in future spondyloarthritis trials.

Development of a technique for MRI gold-standard direct volumetric measurement of complex joint effusion, and validation at the hip

OBJECTIVE

Accurate joint fluid quantification on MRI cannot simply rely on measuring the maximum fluid depth or using an ellipsoid approximation as this does not fully characterize the complex shape of a fluid-filled joint. As per the Outcome Measurement in Rheumatology (OMERACT) filter, we sought to evaluate the feasibility, reliability, and validity of a semi-automated supervised technique to quantify hip effusion volume.

MATERIALS AND METHODS

Ninety-three hip osteoarthritis patients were imaged with coronal short TI inversion recovery (STIR) and sagittal intermediate weighted fat-suppressed (IWFS) sequences at two time points (Fig. 1). Volumetric quantitative measurement (VQM) of joint fluid and measurement of the largest femoral neck fluid thickness (FTM) was performed using the custom MATLAB software. Self-reported Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) and clinical measures of pain, stiffness, and function were recorded.

RESULTS

Inter-observer reliability was significantly higher for VQM than FTM (ICC = 0.96 vs. 0.85, p < 0.05). VQM and FTM correlated moderately (r = 0.76, p < 0.0001). There was significantly more articular fluid in symptomatic than asymptomatic hips at baseline (mean = 9.8 vs. 5.9 mL). Volumetric quantitative measurement generally displayed more frequent and stronger correlations to clinical parameters than FTM. Volumetric quantitative measurement required 3.9 min/hip vs. < 1 min/hip for femoral neck fluid thickness.

CONCLUSION

Volumetric quantitative measurement of joint effusion can serve as an MRI gold-standard, could apply to other joints and collections, and is highly suited to future automation.

Volumetric quantitative measurement of hip effusions by manual versus automated artificial intelligence techniques: An OMERACT preliminary validation study

OBJECTIVE

Preliminary assessment, via OMERACT filter, of manual and automated MRI hip effusion Volumetric Quantitative Measurement (VQM).

METHODS

For 358 hips (93 osteoarthritis subjects, bilateral, 2 time points), 2 radiologists performed manual VQM using custom Matlab software. A Mask R-CNN artificial-intelligence (AI) tool was trained to automatically compute joint fluid volumes.

RESULTS

Manual VQM had excellent inter-observer reliability (ICC 0.96). AI predicted hip fluid volumes with ICC 0.86 (status), 0.58 (change) vs. 2 human readers.

CONCLUSION

Hip joint fluid volumes are reliably assessed by VQM. It is feasible to automate this approach using AI, with promising initial reliability.

Impact of scan quality on AI assessment of hip dysplasia ultrasound

AIMS

Early diagnosis of developmental dysplasia of the hip (DDH) using ultrasound (US) is safe, effective and inexpensive, but requires high-quality scans. The effect of scan quality on diagnostic accuracy is not well understood, especially as artificial intelligence (AI) begins to automate such diagnosis. In this paper, we developed a 10-point scoring system for reporting DDH US scan quality, evaluated its inter-rater agreement and examined its effect on automated assessment by an AI system-MEDO-Hip.

METHODS

Scoring was based on iliac wing straightness and angulation; visibility of labrum, os ischium and femoral head; motion; and other artifacts. Four readers from novice to expert separately scored the quality of 107 scans with this 10-point scale and with holistic grading on a scale of 1-5. MEDO-Hip interpreted the same scans, providing a diagnostic category or identifying the scan as uninterpretable.

RESULTS

Inter-rater agreement for the 10-point scale was significantly higher than holistic scoring ICC 0.68 vs 0.93, p < 0.05. Inter-rater agreement on the categorisation of individual features, by Cohen's kappa, was highest for os ischium (0.67 ± 0.06), femoral head (0.65 ± 0.07) and iliac wing (0.49 ± 0.12) indices, and lower for the presence of labrum (0.21 ± 0.19). MEDO-Hip interpreted all images of a quality > 7 and flagged 13/107 as uninterpretable. These were low-quality images (3 ± 1.2 vs. 7 ± 1.8 in others, p < 0.05), with poor visualization of the os ischium and noticeable motion. AI accuracy in cases with quality scores <  = 7 was 57% vs. 89% on other cases, p < 0.01.

CONCLUSION

This study validates that our scoring system reliably characterises scan quality, and identifies cases likely to be misinterpreted by AI. This could lead to more accurate use of AI in DDH diagnosis by flagging low-quality scans likely to provide poor diagnosis up front.

Discrete Choice Experiment on a Magnetic Resonance Imaging Scoring System for Temporomandibular Joints in Juvenile Idiopathic Arthritis

OBJECTIVE

To determine the relative importance weights of items and grades of a newly developed additive outcome measure called the juvenile idiopathic arthritis (JIA) magnetic resonance imaging (MRI) scoring system for temporomandibular joints (TMJ, JAMRIS-TMJ).

METHODS

An adaptive partial-profile discrete choice experiment (DCE) survey using the 1000Minds platform was independently completed by members of an expert group consisting of radiologists and non-radiologist clinicians to determine the group-averaged relative weights for JAMRIS-TMJ. Subsequently, an image-based vignette ranking exercise was done, during which experts individually rank-ordered 14 patient vignettes for disease severity while blinded to the weights and unrestricted to JAMRIS-TMJ assessment criteria. Validity of the weighted JAMRIS-TMJ was tested by comparing the consensus-graded, DCE-weighted JAMRIS-TMJ score of the vignettes with their unrestricted image-based ranks provided by the experts.

RESULTS

Nineteen experts completed the DCE survey and 21 completed the vignette ranking exercise. Synovial thickening and joint enhancement showed higher weights per raw score compared to bone marrow items and effusion in the inflammatory domain, while erosions and condylar flattening showed non-linear and higher weights compared to disk abnormalities in the damage domain. The weighted JAMRIS-TMJ score of the vignettes correlated highly with the ranks from the unrestricted comparison method, with median Spearman's rho of 0.92 (intra-quartile range: 0.87-0.95) for the inflammation and 0.93 (0.90-0.94) for the damage domain.

CONCLUSIONS

A DCE survey was used to quantify the importance weights of the items and grades of the JAMRIS-TMJ. The weighted score showed high convergent validity with an unrestricted, holistic vignette ranking method.