Vision transformer assisting rheumatologists in screening for capillaroscopy changes in systemic sclerosis: an artificial intelligence model.

OBJECTIVES: The first objective of this study was to implement and assess the performance and reliability of a vision transformer (ViT)-based deep-learning model, an 'off-the-shelf' artificial intelligence solution, for identifying distinct signs of microangiopathy in nailfold capilloroscopy (NFC) images of patients with SSc. The second objective was to compare the ViT's analysis performance with that of practising rheumatologists. METHODS: NFC images of patients prospectively enrolled in our European Scleroderma Trials and Research group (EUSTAR) and Very Early Diagnosis of Systemic Sclerosis (VEDOSS) local registries were used. The primary outcome investigated was the ViT's classification performance for identifying disease-associated changes (enlarged capillaries, giant capillaries, capillary loss, microhaemorrhages) and the presence of the scleroderma pattern in these images using a cross-fold validation setting. The secondary outcome involved a comparison of the ViT's performance vs that of rheumatologists on a reliability set, consisting of a subset of 464 NFC images with majority vote-derived ground-truth labels. RESULTS: We analysed 17 126 NFC images derived from 234 EUSTAR and 55 VEDOSS patients. The ViT had good performance in identifying the various microangiopathic changes in capillaries by NFC [area under the curve (AUC) from 81.8% to 84.5%]. In the reliability set, the rheumatologists reached a higher average accuracy, as well as a better trade-off between sensitivity and specificity compared with the ViT. However, the annotators' performance was variable, and one out of four rheumatologists showed equal or lower classification measures compared with the ViT. CONCLUSIONS: The ViT is a modern, well-performing and readily available tool for assessing patterns of microangiopathy on NFC images, and it may assist rheumatologists in generating consistent and high-quality NFC reports; however, the final diagnosis of a scleroderma pattern in any individual case needs the judgement of an experienced observer.

Self-monitoring of the resting heart rate using a fitness tracker smartwatch application leads to an early diagnosis of large vessel vasculitis.

Giant cell arteritis can involve both cranial and extracranial arteries. Isolated extracranial large vessel vasculitis more often manifests with non-specific constitutional symptoms, causing a diagnostic delay. We report the case of a 57-year-old Caucasian female patient presenting with persistently elevated resting heart rate, as revealed by a smartwatch healthcare application, and non-specific constitutional symptoms. Imaging revealed inflammation of the aorta, bilateral subclavian and axillary arteries, compatible with large vessel vasculitis. Treatment with glucocorticoids and tocilizumab led to a significant improvement of her symptoms and decrease in inflammatory parameters. In sum, an unexplained elevated resting heart rate may lead to an earlier diagnosis and treatment of large vessel vasculitis, especially when other manifestations are non-specific. The use of healthcare smartwatch applications may prove useful in the future and lead to an earlier referral of patients to a physician.

Predictors of mortality of influenza virus infections in a Swiss Hospital during four influenza seasons: Role of quick sequential organ failure assessment.

BACKGROUND: Influenza infections have been associated with high morbidity. The aims were to determine predictors of mortality among patients with influenza infections and to ascertain the role of quick Sequential Organ Failure Assessment (qSOFA) in predicting poor outcomes. METHODS: All adult patients with influenza infection at the Hospital of Jura, Switzerland during four influenza seasons (2014/15 to 2017/18) were included. Cepheid Xpert Xpress Flu/RSV was used during the first three influenza seasons and Cobas Influenza A/B and RSV during the 2017/18 season. RESULTS: Among 1684 influenza virus tests performed, 441 patients with influenza infections were included (238 for influenza A virus and 203 for B). The majority of infections were community onset (369; 83.7%). Thirty-day mortality was 6.0% (25 patients). Multivariate analysis revealed that infection due to A virus (P 0.035; OR 7.1; 95% CI 1.1-43.8), malnutrition (P < 0.001; OR 25.0; 95% CI 4.5-138.8), hospital-acquired infection (P 0.003; OR 12.2; 95% CI 2.3-65.1), respiratory insufficiency (PaO2/FiO2 < 300) (P < 0.001; OR 125.8; 95% CI 9.6-1648.7) and pulmonary infiltrate on X-ray (P 0.020; OR 6.0; 95% CI 1.3-27.0) were identified as predictors of mortality. qSOFA showed a very good accuracy (0.89) equivalent to other more specific and burdensome scores such as CURB-65 and Pneumonia Severity Index (PSI). CONCLUSION: qSOFA performed similarly to specific severity scores (PSI, CURB-65) in predicting mortality. Infection by influenza A virus, respiratory insufficiency and malnutrition were associated with worse prognosis.