Standardized reporting of spine and sacroiliac joints in axial spondyloarthritis MRI: from the ESSR-Arthritis Subcommittee.

OBJECTIVES: Apply a modified Delphi-based approach and produce a practical, radiology-specific set of definitions for interpretation and standardization of the multiple MRI findings in axial spondyloarthritis (ax-SpA), specifically to aid the general radiologist with a musculoskeletal interest, working with gold standard basic MRI protocols. MATERIALS AND METHODS: We report the results of a modified Delphi-based consensus of 35 experts from 13 countries in the Arthritis Subcommittee of the European Society of Musculoskeletal Radiology (ESSR). Seventeen definitions were created (i.e., nine for the spine and eight for the sacroiliac joint) and two Delphi rounds were conducted on an electronic database, collated and revised by the project leader with agreement. Group leads were appointed for each definition following the first round. Final definitions included only those that reached a consensus > 80%; if > 50% agreed on exclusion consensus, definitions were excluded. Final results have been shared during the Arthritis meeting at the Annual ESSR Congress. RESULTS: Fourteen definitions, eight for the spine and six for the sacroiliac joint were agreed for standardized reporting. Andersson's, anterior corner sclerotic and costovertebral joint inflammatory lesions of the spine, with active and non-active erosions, and fat metaplasia of the sacroiliac joint reaching the highest consensus (≥ 95%). More than 50% of the experts agreed to exclude joint space inflammation in the sacroiliac joint and tissue backfill. Syndesmophytes reached 76% agreement. CONCLUSIONS: Agreed definitions by expert radiologists using a modified Delphi process, should allow standardized actionable radiology reports and clarity in reporting terminology of ax-SpA. CLINICAL RELEVANCE STATEMENT: The proposed definitions will support reporting from musculoskeletal and general radiologists working with gold-standard basic MRI, improve confidence in lesion assessment, and standardize terminology to provide actionable reports on MRI in patients with ax-SpA. KEY POINTS: Experts applied a modified Delphi method to optimize the definitions of MRI findings of ax-SpA. After two Delphi rounds and one in-person meeting, fourteen definitions reached the agreement threshold. These consensus-based definitions will aid in actionable reporting specifically for the general radiologist with a musculoskeletal interest.

Patellar Instability: Imaging Findings.

Patellofemoral instability results from impaired engagement of the patella in the trochlear groove at the start of flexion and may lead to pain and lateral patellar dislocation. It occurs most frequently in adolescents and young adults during sporting activities. Trochlear dysplasia, patella alta, and excessive lateralization of the tibial tuberosity are the most common risk factors for patellar instability. The main role of imaging is to depict and assess these anatomical factors and highlight features indicating previous lateral dislocation of the patella.

Commercially-available AI algorithm improves radiologists' sensitivity for wrist and hand fracture detection on X-ray, compared to a CT-based ground truth.

OBJECTIVES: Algorithms for fracture detection are spreading in clinical practice, but the use of X-ray-only ground truth can induce bias in their evaluation. This study assessed radiologists' performances to detect wrist and hand fractures on radiographs, using a commercially-available algorithm, compared to a computerized tomography (CT) ground truth. METHODS: Post-traumatic hand and wrist CT and concomitant X-ray examinations were retrospectively gathered. Radiographs were labeled based on CT findings. The dataset was composed of 296 consecutive cases: 118 normal (39.9%), 178 pathological (60.1%) with a total of 267 fractures visible in CT. Twenty-three radiologists with various levels of experience reviewed all radiographs without AI, then using it, blinded towards CT results. RESULTS: Using AI improved radiologists' sensitivity (Se, 0.658 to 0.703, p < 0.0001) and negative predictive value (NPV, 0.585 to 0.618, p < 0.0001), without affecting their specificity (Sp, 0.885 vs 0.891, p = 0.91) or positive predictive value (PPV, 0.887 vs 0.899, p = 0.08). On the radiographic dataset, based on the CT ground truth, stand-alone AI performances were 0.771 (Se), 0.898 (Sp), 0.684 (NPV), 0.915 (PPV), and 0.764 (AUROC) which were lower than previously reported, suggesting a potential underestimation of the number of missed fractures in the AI literature. CONCLUSIONS: AI enabled radiologists to improve their sensitivity and negative predictive value for wrist and hand fracture detection on radiographs, without affecting their specificity or positive predictive value, compared to a CT-based ground truth. Using CT as gold standard for X-ray labels is innovative, leading to algorithm performance poorer than reported elsewhere, but probably closer to clinical reality. CLINICAL RELEVANCE STATEMENT: Using an AI algorithm significantly improved radiologists' sensitivity and negative predictive value in detecting wrist and hand fractures on radiographs, with ground truth labels based on CT findings. KEY POINTS: • Using CT as a ground truth for labeling X-rays is new in AI literature, and led to algorithm performance significantly poorer than reported elsewhere (AUROC: 0.764), but probably closer to clinical reality. • AI enabled radiologists to significantly improve their sensitivity (+ 4.5%) and negative predictive value (+ 3.3%) for the detection of wrist and hand fractures on X-rays. • There was no significant change in terms of specificity or positive predictive value.

Crystal Diseases of the Spine.

Gout, calcium pyrophosphate deposition disease, and apatite calcifications, the three main crystal disorders, may involve the spine. These disorders can be completely asymptomatic or associated with various clinical symptoms, such as acute flares and more chronic manifestations. This article presents the typical and more unusual imaging features encountered in these disorders.

Relationship between bone marrow adipose tissue and kidney function in postmenopausal women.

Introduction: Bone marrow adipose tissue (BMAT) is associated with aging, osteoporosis, and chronic kidney disease (CKD). To date, the association between BMAT and kidney function in postmenopausal women has not been thoroughly investigated. The main purpose of this study was to determine whether a relationship exists between proton density fat fraction (PDFF) and kidney function in postmenopausal women. Methods: We investigated the cross-sectional association between estimated glomerular filtration rate (eGFR) - calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation - and PDFF - measured at the lumbar spine and proximal femur using Water Fat Imaging (WFI) MRI - in 199 postmenopausal women from the ADIMOS cohort study. We also performed DXA scans and laboratory measurements of sclerostin and c-terminal Fibroblast Growth Factor 23 (cFGF23). Results: Participants' mean age was 67.5 (standard deviation, SD 10.0) years. Their median eGFR was 85.0 (interquartile range, IQR 72.2-95.0) ml/min/1.73 cm2, and their mean lumbar spine PDFF was 57.9 % (SD 9.6). When classified by eGFR-based CKD stages, 41.7 % of the cohort had an eGFR ≥ 90 (n = 83), 47.2 % had an eGFR of 60-89.9 (n = 94), and 11.1 % had an eGFR of 30-59.9 (n = 22). Participants with eGFR ≥ 90 had a lower lumbar spine PDFF than those with eGFR 60-89.9 (mean 55.8 % (9.8) vs. 58.9 % (9.0), p = 0.031) and those with eGFR 30-59.9 (55.8 % (9.8) vs. 60.8 % (9.8), p = 0.043). However, the differences did not remain significant after adjusting for predetermined confounders, including age, diabetes, Charlson comorbidity index, recent history of fragility fracture, appendicular lean mass, and lumbar spine BMD. The inclusion of sclerostin and/or cFGF23 as suspected mediators did not alter the findings. When proximal hip imaging-based PDFF was considered, no significant differences were found between the eGFR categories in the unadjusted and adjusted analyses. Conclusion: No evidence of an association between kidney function and bone marrow adiposity was found either in the lumbar spine or proximal femur in a cohort of postmenopausal women.

Metaphyseal and Diaphyseal Contours: Variants and Pitfalls.

We discuss several variants of the metaphyseal and diaphyseal bone surfaces that may be misleading in clinical practice. They include metaphyseal stripes, spiculated metaphyseal cortex, cortical desmoid, laminated lateral supracondylar ridge, cortical vascular canals, variations in shape or lucency of normal tuberosities, cortical thickening of normal ridges, and well-organized undulated hyperostosis at the proximal phalanges.

Myosteatosis and bone marrow adiposity are not associated among postmenopausal women with fragility fractures.

Objectives: Although paravertebral intramuscular fatty infiltration (known as myosteatosis) following a vertebral fracture is well-known, scarce data are available regarding interactions between muscle, bone, and other fat depots. Based on a homogeneous cohort comprising postmenopausal women with or without a history of fragility fracture, we aimed to better depict the interrelationship between myosteatosis and bone marrow adiposity (BMA). Methods: 102 postmenopausal women were included, 56 of whom had a fragility fracture. Mean proton density fat fraction (PDFF) was measured in the psoas (PDFFPsoas) and paravertebral (PDFFParavertebral) muscles at the lumbar level, as well as in the lumbar spine and non-dominant hip using chemical shift encoding-based water-fat imaging. Visceral adipose tissue (VAT) and total body fat (TBF) were assessed using dual X-ray absorptiometry. Statistical models were adjusted for age, weight, height (all comparisons), and bone mineral density (when considering BMA). Results: PDFF in the psoas and paravertebral muscles was higher in the fracture group compared to controls even after adjustment for age, weight, and height (PDFFPsoas = 17.1 ± 6.1% versus 13.5 ± 4.9%, p=0.004; PDFFParavertebral = 34.4 ± 13.6% versus 24.9 ± 8.8%, p=0.002). Higher PDFFParavertebral was associated with lower PDFF at the lumbar spine (ß = -6.80 ± 2.85, p=0.022) among controls but not in the fracture group. In both groups, a significant relationship between higher PDFFPsoas and higher VAT was observed (ß = 20.27 ± 9.62, p=0.040 in the fracture group, and ß = 37.49 ± 8.65, p<0.001 in the control group). Although solely observed among controls, a similar relationship was observed between PDFFParavertebral and TBF (ß = 6.57 ± 1.80, p<0.001). No significant association was observed between BMA and other fat depots. Conclusion: Myosteatosis is not associated with BMA among postmenopausal women with fragility fractures. Whereas myosteatosis was associated with other fat depots, BMA appears uniquely regulated.

Detection and quantification of pulmonary embolism with artificial intelligence: The SFR 2022 artificial intelligence data challenge.

PURPOSE: In 2022, the French Society of Radiology together with the French Society of Thoracic Imaging and CentraleSupelec organized their 13th data challenge. The aim was to aid in the diagnosis of pulmonary embolism, by identifying the presence of pulmonary embolism and by estimating the ratio between right and left ventricular (RV/LV) diameters, and an arterial obstruction index (Qanadli's score) using artificial intelligence. MATERIALS AND METHODS: The data challenge was composed of three tasks: the detection of pulmonary embolism, the RV/LV diameter ratio, and Qanadli's score. Sixteen centers all over France participated in the inclusion of the cases. A health data hosting certified web platform was established to facilitate the inclusion process of the anonymized CT examinations in compliance with general data protection regulation. CT pulmonary angiography images were collected. Each center provided the CT examinations with their annotations. A randomization process was established to pool the scans from different centers. Each team was required to have at least a radiologist, a data scientist, and an engineer. Data were provided in three batches to the teams, two for training and one for evaluation. The evaluation of the results was determined to rank the participants on the three tasks. RESULTS: A total of 1268 CT examinations were collected from the 16 centers following the inclusion criteria. The dataset was split into three batches of 310, 580 and 378 C T examinations provided to the participants respectively on September 5, 2022, October 7, 2022 and October 9, 2022. Seventy percent of the data from each center were used for training, and 30% for the evaluation. Seven teams with a total of 48 participants including data scientists, researchers, radiologists and engineering students were registered for participation. The metrics chosen for evaluation included areas under receiver operating characteristic curves, specificity and sensitivity for the classification task, and the coefficient of determination r2 for the regression tasks. The winning team achieved an overall score of 0.784. CONCLUSION: This multicenter study suggests that the use of artificial intelligence for the diagnosis of pulmonary embolism is possible on real data. Moreover, providing quantitative measures is mandatory for the interpretability of the results, and is of great aid to the radiologists especially in emergency settings.

Interdisciplinary consensus statements on imaging of DRUJ instability and TFCC injuries.

OBJECTIVES: The purpose of this agreement was to establish evidence-based consensus statements on imaging of distal radioulnar joint (DRUJ) instability and triangular fibrocartilage complex (TFCC) injuries by an expert group using the Delphi technique. METHODS: Nineteen hand surgeons developed a preliminary list of questions on DRUJ instability and TFCC injuries. Radiologists created statements based on the literature and the authors' clinical experience. Questions and statements were revised during three iterative Delphi rounds. Delphi panelists consisted of twenty-seven musculoskeletal radiologists. The panelists scored their degree of agreement to each statement on an 11-item numeric scale. Scores of "0," "5," and "10" reflected complete disagreement, indeterminate agreement, and complete agreement, respectively. Group consensus was defined as a score of "8" or higher for 80% or more of the panelists. RESULTS: Three of fourteen statements achieved group consensus in the first Delphi round and ten statements achieved group consensus in the second Delphi round. The third and final Delphi round was limited to the one question that did not achieve group consensus in the previous rounds. CONCLUSIONS: Delphi-based agreements suggest that CT with static axial slices in neutral rotation, pronation, and supination is the most useful and accurate imaging technique for the work-up of DRUJ instability. MRI is the most valuable technique in the diagnosis of TFCC lesions. The main indication for MR arthrography and CT arthrography are Palmer 1B foveal lesions of the TFCC. CLINICAL RELEVANCE STATEMENT: MRI is the method of choice for assessing TFCC lesions, with higher accuracy for central than peripheral abnormalities. The main indication for MR arthrography is the evaluation of TFCC foveal insertion lesions and peripheral non-Palmer injuries. KEY POINTS: • Conventional radiography should be the initial imaging technique in the assessment of DRUJ instability. CT with static axial slices in neutral rotation, pronation, and supination is the most accurate method for evaluating DRUJ instability. • MRI is the most useful technique in diagnosing soft-tissue injuries causing DRUJ instability, especially TFCC lesions. • The main indications for MR arthrography and CT arthrography are foveal lesions of the TFCC.

Anatomical Variants of the Sacroiliac Joint.

Anatomical variants are frequently encountered when assessing the sacroiliac joints (SIJ) using magnetic resonance imaging. When not located in the weight-bearing part of the SIJ, variants associated with structural and edematous changes can be misinterpreted as sacroiliitis. Their correct identification is necessary to avoid radiologic pitfalls. This article reviews five SIJ variants involved in the dorsal ligamentous space (accessory SIJ, iliosacral complex, semicircular defect, bipartite iliac bony plate, and crescent iliac bony plate) and three SIJ variants involved in the cartilaginous part of the SIJ (posterior dysmorphic SIJ, isolated synostosis, and unfused ossification centers).

Bone Marrow Adiposity and Fragility Fractures in Postmenopausal Women: The ADIMOS Case-Control Study.

CONTEXT: Noninvasive assessment of proton density fat fraction (PDFF) by magnetic resonance imaging (MRI) may improve the prediction of fractures. OBJECTIVE: This work aimed to determine if an association exists between PDFF and fractures. METHODS: A case-control study was conducted at Lille University Hospital, Lille, France, with 2 groups of postmenopausal women: one with recent osteoporotic fractures, and the other with no fractures. Lumbar spine and proximal femur (femoral head, neck, and diaphysis) PDFF were determined using chemical shift-based water-fat separation MRI (WFI) and dual-energy x-ray absorptiometry scans of the lumbar spine and hip. Our primary objective was to determine the relationship between lumbar spine PDFF and osteoporotic fractures in postmenopausal women. Analysis of covariance was used to compare PDFF measurements between patient cases (overall and according to the type of fracture) and controls, after adjusting for age, Charlson comorbidity index (CCI) and BMD. RESULTS: In 199 participants, controls (n = 99) were significantly younger (P < .001) and had significantly higher BMD (P < 0.001 for all sites) than patient cases (n = 100). A total of 52 women with clinical vertebral fractures and 48 with nonvertebral fractures were included. When PDFFs in patient cases and controls were compared, after adjustment on age, CCI, and BMD, no statistically significant differences between the groups were found at the lumbar spine or proximal femur. When PDFFs in participants with clinical vertebral fractures (n = 52) and controls were compared, femoral neck PDFF and femoral diaphysis PDFF were detected to be lower in participants with clinical vertebral fractures than in controls (adjusted mean [SE] 79.3% [1.2] vs 83.0% [0.8]; P = 0.020, and 77.7% [1.4] vs 81.6% [0.9]; P = 0.029, respectively). CONCLUSION: No difference in lumbar spine PDFF was found between those with osteoporotic fractures and controls. However, imaging-based proximal femur PDFF may discriminate between postmenopausal women with and without clinical vertebral fractures, independently of age, CCI, and BMD.

Generative adversarial networks (GAN)-based data augmentation of rare liver cancers: The SFR 2021 Artificial Intelligence Data Challenge.

PURPOSE: The 2021 edition of the Artificial Intelligence Data Challenge was organized by the French Society of Radiology together with the Centre National d'Études Spatiales and CentraleSupélec with the aim to implement generative adversarial networks (GANs) techniques to provide 1000 magnetic resonance imaging (MRI) cases of macrotrabecular-massive (MTM) hepatocellular carcinoma (HCC), a rare and aggressive subtype of HCC, generated from a limited number of real cases from multiple French centers. MATERIALS AND METHODS: A dedicated platform was used by the seven inclusion centers to securely upload their anonymized MRI examinations including all three cross-sectional images (one late arterial and one portal-venous phase T1-weighted images and one fat-saturated T2-weighted image) in compliance with general data protection regulation. The quality of the database was checked by experts and manual delineation of the lesions was performed by the expert radiologists involved in each center. Multidisciplinary teams competed between October 11th, 2021 and February 13th, 2022. RESULTS: A total of 91 MTM-HCC datasets of three images each were collected from seven French academic centers. Six teams with a total of 28 individuals participated in this challenge. Each participating team was asked to generate one thousand 3-image cases. The qualitative evaluation was performed by three radiologists using the Likert scale on ten randomly selected cases generated by each participant. A quantitative evaluation was also performed using two metrics, the Frechet inception distance and a leave-one-out accuracy of a 1-Nearest Neighbor algorithm. CONCLUSION: This data challenge demonstrates the ability of GANs techniques to generate a large number of images from a small sample of imaging examinations of a rare malignant tumor.

Anatomical study of the dorsal capsulo-scapholunate septum using high frequency ultrasonography.

INTRODUCTION: The dorsal capsulo-scapholunate septum (DCSS) is a recently described capsuloligamentous structure between the dorsal bundle of the scapholunate ligament (SLL) and the joint capsule of the wrist. It acts a secondary stabilizer of the scapholunate joint. The aim of this study was to evaluate the visibility and normal appearance of DCSS on high frequency ultrasound. HYPOTHESIS: The DCSS can be analyzed using high frequency ultrasound. MATERIALS AND METHODS: Three cadaveric wrists were dissected in order to study the DCSS; one without labeling and the other two after labeling under ultrasound guidance. On two other wrists, a correlation between the structure considered to be the DCSS on ultrasound and the corresponding CT arthrography and anatomical sections was carried out. Finally, sagittal ultrasound sections of the DCSS region on 42 healthy wrists were analyzed retrospectively. RESULTS: During dissection, the DCSS corresponded to a fibrous structure extending to the dorsal surface of the scaphoid and lunate, with certain fibers converging towards the SLL. On high-frequency ultrasound, a hyperechoic fibrillar structure was visualized at the theoretical position of the DCSS. The dissections performed after ultrasound-guided transfixion showed that its limits corresponded to the limits of the DCSS. The anatomical, CT arthrography and ultrasound sections in the DCSS region were concordant. The DCSS was retrospectively visible on at least one reference ultrasound slice in 90.5% of healthy wrists, and was always hyperechoic. Its average thickness was 1.38±0.24mm. DISCUSSION: Our study describes the normal sonographic characteristics of the DCSS, visible in a majority of healthy patients on high-frequency ultrasound. The analysis of the sonographic characteristics of the DCSS in the context of acute wrist trauma should be evaluated, and a diagnostic decision tree has been proposed. LEVEL OF EVIDENCE: IV.

Quantitative dynamic contrast-enhanced MRI of bone marrow perfusion at the proximal femur: influence of femoral head osteonecrosis risk factor and overt osteonecrosis.

OBJECTIVE: To evaluate the MRI perfusion changes in patients with risk factors for osteonecrosis and normally appearing femoral heads prior to overt femoral head osteonecrosis. METHODS: Fifty-eight patients (105 hips) were prospectively included in this ethics committee-approved study. There were 46 hips with no image anomalies and no risk factors for osteonecrosis of the femoral head (ONFH) risk factors, 38 with ONFH risk factors and no image abnormalities, and 21 with overt ONFH. All patients underwent DCE-MRI. Semi-quantitative (peak enhancement [PE], area under the curve [AUC], time to maximum enhancement [TME]) and quantitative perfusion parameters (volume plasma, KTRANS, and KEP) were calculated. Excessive alcohol consumption, corticosteroid use, and trauma were considered major risk factors for osteonecrosis of the femoral head. RESULTS: Measured at the femoral neck and compared to the healthy hips without OFNH risk factors, PE was significantly lower in the hips of patients with OFNH risk factors. Moreover, the difference was greater in females with risk factors, who presented significantly lower PE values (p = 0.0096). A PE threshold of 1.4% yielded a 92% sensitivity and 54% specificity for the presence of associated ONFH risk factors. The hips with overt OFNH compared to those with normally appearing showed an increase of PE of 45% in the neck (p < 0.014). Various epiphyseal femoral head perfusion parameters (PE, TME, AUC, and Ktrans) presented statistically significant differences in hips with ONFH and those without (p < 0.0001). CONCLUSION: DCE-MRI can identify perfusion marrow changes related to the presence of ONFH risk factors and adjacent to osteonecrosis areas. KEY POINTS: • Bone marrow perfusion changes may occur prior to overt ONFH and extend beyond the osteonecrosis area to the entire femoral head and neck. • Peak enhancement values were significantly reduced in patients with ONFH risk factors, compared to those without. • The presence of ONFH led to a significant increase in marrow perfusion adjacent to the osteonecrosis area.

Anatomical study of the sternoclavicular joint using high-frequency ultrasound.

OBJECTIVES: The purpose of the present study was to determine whether ultrasound enables assessment of sternoclavicular structures. METHODS: A preliminary study in 3 cadavers was followed by an ultrasound study, performed by 2 musculoskeletal radiologists working in consensus, in 59 patients without history of trauma, surgery or pain in the sternoclavicular joint. The visibility, echogenicity and thickness of the sternoclavicular structures were assessed. RESULTS: The anterior sternoclavicular ligament and the interclavicular ligament could be seen in all patients (mean thickness: 1.4 mm and 1.3 mm, respectively). The articular disc was clearly seen in 66.1% of cases, and shoulder antepulsion enabled analysis in an additional 20.3%. Intra-articular joint gas was frequent (33.89% of cases), preventing analysis of the disc in 2 patients. Only the superficial anterior aspect of the clavicular and sternal articular cartilages could be assessed. Joint effusion was seen in 6.8% of cases. Clavicular osteophytes, sternal osteophytes and bone irregularities at the anterior sternoclavicular ligament insertion were detected in 33.9%, 16.9% and 16.9% of cases, respectively. CONCLUSION: The anterior sternoclavicular ligament, interclavicular ligament and anterior intra-articular structures can be visualized by ultrasound. This means of assessment may have clinical applications, particularly in patients with trauma or microtrauma.

Presurgical Perspective and Postsurgical Evaluation of Non-Achilles Tendons of the Ankle and Retinaculum.

The vast majority of non-Achilles ankle tendinopathies are related to overuse. This article discusses the clinical aspect, imaging appearance, and management of tendinopathies of the lateral, medial, and the anterior compartments with a focus on presurgical perspective and postsurgical evaluation.

Minimally invasive removal of deep contraceptive implants under continuous ultrasound guidance is effective, quick, and safe.

OBJECTIVES: The aim of this study was to assess the feasibility, performance, and complications of a non-surgical, minimally-invasive procedure of deep contraceptive implant removal under continuous ultrasound guidance. METHODS: The ultrasound-guided procedure consisted of local anesthesia using lidocaine chlorhydrate 1% (10 mg/mL) with a 21-G needle, followed by hydrodissection using NaCl 0.9% (9 mg/mL) and implant extraction using a Hartmann grasping microforceps. The parameters studied were the implant localization, success and complication rates, pain throughout the intervention, volumes of lidocaïne and NaCl used, duration of the procedure, and size of the incision. Between November 2019 and January 2021, 45 patients were referred to the musculoskeletal radiology department for ultrasound-guided removal of a deep contraceptive implant and were all retrospectively included. RESULTS: All implants were successfully removed en bloc (100%). The mean incision size was 2.7 ± 0.5 mm. The mean duration of the extraction procedure was 7.7 ± 6.3 min. There were no major complications (infection, nerve, or vessel damage). As a minor complication, 21 patients (46.7%) reported a benign superficial skin ecchymosis at the puncture site, spontaneously regressing in less than 1 week. The procedure was very well-tolerated, with low pain rating throughout (1.0 ± 1.5/10 during implant extraction). CONCLUSIONS: Minimally invasive removal of deep contraceptive implants under continuous ultrasound guidance alone is feasible, effective, and safe. In the present cohort, all implants were successfully removed, whatever the location, with short procedural time, small incision size, low pain levels, and no significant complications. This procedure could become a gold standard in this indication. KEY POINTS: • Minimally invasive removal of deep contraceptive implants under continuous ultrasound guidance alone is feasible, which led to a success rate of 100% whatever the location (even close to neurovascular structures), with only a small skin incision (2.7 ± 0.5 mm). • The procedure was safe, quick, without any major complications, and very well tolerated in terms of pain. • This minimally invasive ultrasound-guided procedure could become the future gold standard for the removal of deep contraceptive implants, as an alternative to surgical extraction, even for implants in difficult locations such as subfascial ones or those close to neurovascular structures.

Main Diagnostic Pitfalls in Reading the Sacroiliac Joints on MRI.

Magnetic resonance imaging of the sacroiliac joints is now frequently performed to help identify patients with early axial spondyloarthritis. However, differential diagnoses exist and should be recognized. The aim of this article is to review the most frequent differential diagnoses that may mimic inflammatory sacroiliitis in clinical practice.

Three artificial intelligence data challenges based on CT and ultrasound.

PURPOSE: The 2020 edition of these Data Challenges was organized by the French Society of Radiology (SFR), from September 28 to September 30, 2020. The goals were to propose innovative artificial intelligence solutions for the current relevant problems in radiology and to build a large database of multimodal medical images of ultrasound and computed tomography (CT) on these subjects from several French radiology centers. MATERIALS AND METHODS: This year the attempt was to create data challenge objectives in line with the clinical routine of radiologists, with less preprocessing of data and annotation, leaving a large part of the preprocessing task to the participating teams. The objectives were proposed by the different organizations depending on their core areas of expertise. A dedicated platform was used to upload the medical image data, to automatically anonymize the uploaded data. RESULTS: Three challenges were proposed including classification of benign or malignant breast nodules on ultrasound examinations, detection and contouring of pathological neck lymph nodes from cervical CT examinations and classification of calcium score on coronary calcifications from thoracic CT examinations. A total of 2076 medical examinations were included in the database for the three challenges, in three months, by 18 different centers, of which 12% were excluded. The 39 participants were divided into six multidisciplinary teams among which the coronary calcification score challenge was solved with a concordance index > 95%, and the other two with scores of 67% (breast nodule classification) and 63% (neck lymph node calcifications).