Background Deep learning applied to ultrasound (US) can provide a feedback to the sonographer about the correct identification of scanned tissues and allows for faster and standardized measurements. The most frequently adopted parameter for US diagnosis of carpal tunnel syndrome is the increasing of the cross-sectional area (CSA) of the median nerve. Our aim was to develop a deep learning algorithm, relying on convolutional neural networks (CNNs), for the localization and segmentation of the median nerve and the automatic measurement of its CSA on US images acquired at the proximal inlet of the carpal tunnel. Methods Consecutive patients with rheumatic and musculoskeletal disorders were recruited. Transverse US images were acquired at the carpal tunnel inlet, and the CSA was manually measured. Anatomical variants were registered. The dataset consisted of 246 images (157 for training, 40 for validation, and 49 for testing) from 103 patients each associated with manual annotations of the nerve boundary. A Mask R-CNN, state-of-the-art CNN for image semantic segmentation, was trained on this dataset to accurately localize and segment the median nerve section. To evaluate the performances on the testing set, precision (Prec), recall (Rec), mean average precision (mAP), and Dice similarity coefficient (DSC) were computed. A sub-analysis excluding anatomical variants was performed. The CSA was automatically measured by the algorithm. Results The algorithm correctly identified the median nerve in 41/49 images (83.7%) and in 41/43 images (95.3%) excluding anatomical variants. The following metrics were obtained (with and without anatomical variants, respectively): Prec 0.86 ± 0.33 and 0.96 ± 0.18, Rec 0.88 ± 0.33 and 0.98 ± 0.15, mAP 0.88 ± 0.33 and 0.98 ± 0.15, and DSC 0.86 ± 0.19 and 0.88 ± 0.19. The agreement between the algorithm and the sonographer CSA measurements was excellent [ICC 0.97 (0.94–0.98)]. Conclusions The developed algorithm has shown excellent performances, especially if excluding anatomical variants. Future research should aim at expanding the US image dataset including a wider spectrum of normal anatomy and pathology. This deep learning approach has shown very high potentiality for a fully automatic support for US assessment of carpal tunnel syndrome.

Figure 1Vaccination acceptance among patients with autoimmune inflammatory rheumatic diseases (on the left) and main reasons for declining (on the right). All the comparisons (*) are significant (p<0.01). AEs, adverse events.

Objective To explore the association of the OMERACT ultrasound (US) entheseal abnormalities with the presence of US joint bone erosions in psoriatic arthritis (PsA). Methods Consecutive PsA patients were included in this cross-sectional study. Demographic and clinical parameters were collected. A bilateral US assessment was carried out at the following entheses: plantar fascia, quadriceps, patellar (proximal and distal) and Achilles tendons. The following US entheseal abnormalities were registered: hypoechogenicity, thickening, Doppler signal <2mm from the bony cortex, calcification/enthesophyte, bone erosion. The presence of US joint bone erosions was investigated at the 2nd and 5th metacarpophalangeal (MCP) joints, ulnar head and 5th metatarsophalangeal (MTP) joint, bilaterally, as well as at the level of the most inflamed joint on physical examination. Multiple linear regression analysis was performed to identify clinical and/or US variables associated with US-detected joint bone erosions. Results A total of 104 PsA patients were enrolled. At least one joint bone erosion was found in 47/104 patients (45.2%). Bone erosions were most frequently detected at 5th MTP joint level (42/208 joints, 20.2 %; 32/104 patients, 30.8%). In the multivariate model, only PD signal at the enthesis (P<0.001, standardized β=0.51), bone erosions at the enthesis (P=0.02, standardized β=0.2), PsA disease duration (P=0.04, standardized β=0.17) and greyscale joint synovitis (P=0.03, standardized β=0.42) were associated with US-detected joint bone erosions. Conclusion PD signal and bone erosions at the enthesis represent sonographic biomarkers of a more severe subset of PsA in terms of US-detected joint erosive damage.

ObjectivesTo investigate the inter/intra-reliability of ultrasound (US) muscle echogenicity in patients with rheumatic diseases.MethodsForty-two rheumatologists and 2 radiologists from 13 countries were asked to assess US muscle echogenicity of quadriceps muscle in 80 static images and 20 clips from 64 patients with different rheumatic diseases and 8 healthy subjects. Two visual scales were evaluated, a visual semi-quantitative scale (0–3) and a continuous quantitative measurement (“VAS echogenicity,” 0–100). The same assessment was repeated to calculate intra-observer reliability. US muscle echogenicity was also calculated by an independent research assistant using a software for the analysis of scientific images (ImageJ). Inter and intra reliabilities were assessed by means of prevalence-adjusted bias-adjusted Kappa (PABAK), intraclass correlation coefficient (ICC) and correlations through Kendall’s Tau and Pearson’s Rho coefficients.ResultsThe semi-quantitative scale showed a moderate inter-reliability [PABAK = 0.58 (0.57–0.59)] and a substantial intra-reliability [PABAK = 0.71 (0.68–0.73)]. The lowest inter and intra-reliability results were obtained for the intermediate grades (i.e., grade 1 and 2) of the semi-quantitative scale. “VAS echogenicity” showed a high reliability both in the inter-observer [ICC = 0.80 (0.75–0.85)] and intra-observer [ICC = 0.88 (0.88–0.89)] evaluations. A substantial association was found between the participants assessment of the semi-quantitative scale and “VAS echogenicity” [ICC = 0.52 (0.50–0.54)]. The correlation between these two visual scales and ImageJ analysis was high (tau = 0.76 and rho = 0.89, respectively).ConclusionThe results of this large, multicenter study highlighted the overall good inter and intra-reliability of the US assessment of muscle echogenicity in patients with different rheumatic diseases.

Background:Enthesitis is one of the hallmark of psoriatic arthritis (PsA). Ultrasound (US) accurately detects morphostructural abnormalities indicative of entheseal inflammation and structural damage. Interestingly, in a recent study, US-detected entheseal pathology appeared to be a potential marker of disease severity, being associated with higher radiographic score of structural damage at peripheral joint level. (1) However, a sub-analysis of the impact of each elementary finding of US enthesitis was not performed. Moreover, some US entheseal abnormalities (hypoechogenicity, thickening and calcification/enthesophyte) have been described as frequent findings in healthy subjects and patients with dysmetabolic conditions, undermining their specificity. (2) Thus, we hypothesized that their role as a sonographic biomarker of joint disease severity could be questioned.Objectives:The main aim of the present study was to explore the association between the US elementary findings of enthesitis defined by OMERACT [i.e. hypoechogenicity, thickening, Doppler signal, calcification/enthesophyte and bone erosion at enthesis] (3) and the presence of US-detected joint bone erosions in patients with PsA.Methods:Consecutive patients with PsA (CASPAR criteria) were included in this cross-sectional single-centre study. The scanning protocol included bilateral assessment of the main entheses of the lower limbs [plantar fascia, quadriceps, patellar (proximal and distal) and Achilles tendons]. The presence of US joint bone erosions was investigated in the following areas: 2nd and 5th metacarpophalangeal (MCP) joints, ulnar head and 5th metatarsophalangeal (MTP) joint, bilaterally, as well as the most inflamed joint at the physical examination. The US examination was carried out with a 6-18 MHz probe. Univariate and multivariate logistic analysis were performed to identify predictors of US joint bone erosions.Results:A total of 74 PsA patients were enrolled. The mean disease duration was 7.9±8.0 years. Joint bone erosions were found in 36/75 patients (48.0%), and in 71/600 joints (11.8%), most frequently in the 5th MTP joint (in 26/75 patients, 34.7%). The univariate analysis showed that entheseal bone erosions [odds ratio (OR) 27.1, 95% confidence interval (CI) 3.3-220.2, p value <0.01] and Doppler signal (OR 3.5, 95% CI 1.3 - 9.4, p value 0.01) were associated with joint bone erosions. Only entheseal bone erosions remained significantly associated with joint bone erosions in the multivariate analysis (Table 1).Table 1.Multivariate regression analysis: predictive value of the entheseal US findings for the presence of joint bone erosions.OR (95% CI)P valueHypoechogenicity0.5 (0.1-3.4)0.45Thickening2.2 (0.6-8.3)0.27Doppler signal3.2 (0.9-10.8)0.06Calcification/enthesophyte1.1 (0.1-11.2)0.99Entheseal bone erosion24.2 (2.7-216.2)<0.01Conclusion:Entheseal bone erosion and, to a lesser extent, Doppler signal, were the only entheseal abnormalities correlated with the presence of US-detected joint bone erosions, representing potential sonographic biomarkers of disease severity in PsA.References:[1]Polachek A, Cook R, Chandran V, Gladman DD, Eder L. The association between sonographic enthesitis and radiographic damage in psoriatic arthritis. Arthritis Res Ther 2017; 15;19:189.[2]Balint PV, Terslev L, Aegerter P, Bruyn GAW, Chary-Valckenaere I, Gandjbakhch F, et al. Reliability of a consensus-based ultrasound definition and scoring for enthesitis in spondyloarthritis and psoriatic arthritis: An OMERACT US initiative. Ann Rheum Dis 2018; 77(12):1730-5.[3]Di Matteo A, Filippucci E, Cipolletta E, Martire V, Jesus D, Musca A, et al. How normal is the enthesis by ultrasound in healthy subjects? Clin Exp Rheumatol 2020;38:472-8.Disclosure of Interests:Gianluca Smerilli: None declared, Edoardo Cipolletta: None declared, Giulia Maria Destro Castaniti: None declared, Andrea Di Matteo: None declared, Marco Di Carlo: None declared, Erica Moscioni: None declared, Francesca Francioso: None declared, Walter Grassi Speakers bureau: W.G. has received speaking fees from AbbVie, Celgene, Grünenthal, Pfizer and UCB Pharma., Emilio Filippucci Speakers bureau: E.F. has received speaking fees from Abbvie, BMS, Janssen, Lilly, MSD, Novartis, Roche, Pfizer, UCB Pharma.