Characterization of age-associated B cells in early drug-naïve rheumatoid arthritis patients.

Age-associated B cells (ABCs) are an immune cell subset linked to autoimmunity, infection and ageing, and whose pathophysiological importance was recently highlighted using single cell synovial tissue profiling. To elucidate their pathophysiological relevance, peripheral blood (PB) ABCs from early rheumatoid arthritis (eRA) patients naïve to disease-modifying anti-rheumatic drugs (DMARDs) were compared with their synovial fluid (SF) counterparts, and to PB ABCs from psoriatic arthritis patients and healthy controls. PB and SF B-cell subsets were phenotyped by multi-parameter flow cytometry, sorted and subjected to gene expression profiling (NanoString nCounter® Immunology V2 Panel) and functional characterization (stimulated cytokine measurements by immunoassay). PB ABCs of eRA patients, which are transcriptionally distinct from those of control cohorts, express chemokine receptors and adhesion molecules, such as CXCR3, that favour homing to inflammatory sites over lymphoid tissue. These cells are an activated, class-switched B-cell subset expressing high levels of HLA-DR, co-stimulatory molecules and T-bet. Their secretion profile includes IL-12p70 and IL-23 but low levels of IL-10. High surface expression of FcRL family members, including FcRL3, furthermore suggests a role for these cells in autoimmunity. Finally, and unlike in the periphery where they are rare, ABCs are the predominant B-cell subsets in SF. These observations indicate the predilection of ABCs for inflammatory tissue in RA, where their propensity for antigen presentation and pro-inflammatory phenotype may support autoimmune pathology. Their potential as a therapeutic target therefore warrants further study.

Comparison of chest X-ray interpretation by Emergency Department clinicians and radiologists in suspected COVID-19 infection: a retrospective cohort study.

OBJECTIVES: We describe the inter-rater agreement between Emergency Department (ED) clinicians and reporting radiologists in the interpretation of chest X-rays (CXRs) in patients presenting to ED with suspected COVID-19. METHODS: We undertook a retrospective cohort study of patients with suspected COVID-19. We compared ED clinicians' and radiologists' interpretation of the CXRs according to British Society of Thoracic Imaging (BSTI) guidelines, using the area under the receiver operator curve (ROC area). RESULTS: CXRs of 152 cases with suspected COVID-19 infection were included. Sensitivity and specificity for 'classic' COVID-19 CXR findings reported by ED clinician was 84 and 83%, respectively, with a ROC area of 0.84 (95%CI 0.77 to 0.90). Accuracy improved with ED clinicians' experience, with ROC areas of 0.73 (95%CI 0.45 to 1.00), 0.81 (95%CI 0.73 to 0.89), 1.00 (95%CI 1.00 to 1.00) and 0.90 (95%CI 0.70 to 1.00) for foundation year doctors, senior house officers, higher speciality trainees and ED consultants, respectively (p < 0.001). CONCLUSIONS: ED clinicians demonstrated moderate inter-rater agreement with reporting radiologists according to the BSTI COVID-19 classifications. The improvement in accuracy with ED clinician experience suggests training of junior ED clinicians in the interpretation of COVID-19 related CXRs might be beneficial. Large-scale survey studies might be useful in the further evaluation of this topic. ADVANCES IN KNOWLEDGE: This is the first study to examine inter-rater agreement between ED clinicians and radiologists in regards to COVID-19 CXR interpretation.Further service configurations such as 24-hr hot reporting of CXRs can be guided by these data, as well as an ongoing, nationwide follow-up study.

Feasibility of Data-driven EMG Signal Generation using a Deep Generative Model.

Despite recent advancements in the field of pattern recognition-based myoelectric control, the collection of a high quality training set remains a challenge limiting its adoption. This paper proposes a framework for a possible solution by augmenting short training protocols with subject-specific synthetic electromyography (EMG) data generated using a deep generative network, known as SinGAN. The aim of this work is to produce high quality synthetic data that could improve classification accuracy when combined with a limited training protocol. SinGAN was used to generate 1000 synthetic windows of EMG data from a single window of six different motions, and results were evaluated qualitatively, quantitatively, and in a classification task. Qualitative assessment of synthetic data was conducted via visual inspection of principal component analysis projections of real and synthetic feature space. Quantitative assessment of synthetic data revealed 11 of 32 synthetic features had similar location and scale to real features (using univariate two-sample Lepage tests); whereas multivariate distributions were found to be statistically different (p <0.05). Finally, the addition of these synthetic data to a brief training set of real data significantly improved classification accuracy in a cross-validation testing scheme by 5.4% (p <0.001).

Automation of the Timed-Up-and-Go Test Using a Conventional Video Camera.

The Timed-Up-and-Go (TUG) test is a simple clinical tool commonly used to quickly assess the mobility of patients. Researchers have endeavored to automate the test using sensors or motion tracking systems to improve its accuracy and to extract more resolved information about its sub-phases. While some approaches have shown promise, they often require the donning of sensors or the use of specialized hardware, such as the now discontinued Microsoft Kinect, which combines video information with depth sensors (RGBD). In this work, we leverage recent advances in computer vision to automate the TUG test using a regular RGB video camera without the need for custom hardware or additional depth sensors. Thirty healthy participants were recorded using a Kinect V2 and a standard video feed while performing multiple trials of 3 and 1.5 meter versions of the TUG test. A Mask Regional Convolutional Neural Net (R-CNN) algorithm and a Deep Multitask Architecture for Human Sensing (DMHS) were then used together to extract global 3D poses of the participants. The timing of transitions between the six key movement phases of the TUG test were then extracted using heuristic features extracted from the time series of these 3D poses. The proposed video-based vTUG system yielded the same error as the standard Kinect-based system for all six key transitions points, and average errors of less than 0.15 seconds from a multi-observer hand labeled ground truth. This work describes a novel method of video-based automation of the TUG test using a single standard camera, removing the need for specialized equipment and facilitating the extraction of additional meaningful information for clinical use.

B cells expressing the IgA receptor FcRL4 participate in the autoimmune response in patients with rheumatoid arthritis.

The clinical efficacy of B cell targeting therapies highlights the pathogenic potential of B cells in inflammatory diseases. Expression of Fc Receptor like 4 (FcRL4) identifies a memory B cell subset, which is enriched in the joints of patients with rheumatoid arthritis (RA) and in mucosa-associated lymphoid tissue. The high level of RANKL production by this B cell subset indicates a unique pathogenic role. In addition, recent work has identified a role for FcRL4 as an IgA receptor, suggesting a potential function in mucosal immunity. Here, the contribution of FcRL4+ B cells to the specific autoimmune response in the joints of patients with RA was investigated. Single FcRL4+ and FcRL4- B cells were sorted from synovial fluid and tissue from RA patients and their immunoglobulin genes characterized. Levels of hypermutation in the variable regions in both populations were largely consistent with memory B cells selected by an antigen- and T cell-dependent process. Recombinant antibodies were generated based on the IgH and IgL variable region sequences and investigated for antigen specificity. A significantly larger proportion of the recombinant antibodies generated from individual synovial FcRL4+ B cells showed reactivity towards citrullinated autoantigens. Furthermore, both in analyses based on heavy chain sequences and flow cytometric detection, FcRL4+ B cells have significantly increased usage of the IgA isotype. Their low level of expression of immunoglobulin and plasma cell differentiation genes does not suggest current antibody secretion. We conclude that these activated B cells are a component of the local autoimmune response, and through their RANKL expression, can contribute to joint destruction. Furthermore, their expression of FcRL4 and their enrichment in the IgA isotype points towards a potential role for these cells in the link between mucosal and joint inflammation.