Intensive care clinicians are presented with large quantities of patient information and measurements from a multitude of monitoring systems. The limited ability of humans to process such complex information hinders physicians to readily recognize and act on early signs of patient deterioration. We used machine learning to develop an early warning system for circulatory failure based on a high-resolution ICU database with 240 patientyears of data. This automatic system predicts 90.0% of circulatory failure events (prevalence 3.1%), with 81.8% identified more than two hours in advance, resulting in an area under the receiver operating characteristic curve of 94.0% and area under the precision recall curve of 63.0%. The model was externally validated in a large independent patient cohort.

Introduction Since December 2019, a novel coronavirus (SARS-CoV-2) has triggered a world-wide pandemic with an enormous medical and societal-economic toll. Thus, our aim was to gather all available information regarding comorbidities, clinical signs and symptoms, outcomes, laboratory findings, imaging features, and treatments in patients with coronavirus disease 2019 (COVID-19). Methods EMBASE, PubMed/Medline, Scopus, and Web of Science were searched for studies published in any language between December 1st, 2019 and March 28th, 2020. Original studies were included if the exposure of interest was an infection with SARS-CoV-2 or confirmed COVID-19. The primary outcome was the risk ratio of comorbidities, clinical signs and symptoms, laboratory findings, imaging features, treatments, outcomes, and complications associated with COVID-19 morbidity and mortality. We performed random-effects pairwise meta-analyses for proportions and relative risks, I 2 , T 2 , and Cochrane Q, sensitivity analyses, and assessed publication bias. Results 148 studies met the inclusion criteria for the systematic review and meta-analysis with 12′149 patients (5′739 female) and a median age of 47.0 [35.0–64.6] years. 617 patients died from COVID-19 and its complication. 297 patients were reported as asymptomatic. Older age (SMD: 1.25 [0.78–1.72]; p < 0.001), being male (RR = 1.32 [1.13–1.54], p = 0.005) and pre-existing comorbidity (RR = 1.69 [1.48–1.94]; p < 0.001) were identified as risk factors of in-hospital mortality. The heterogeneity between studies varied substantially ( I 2 ; range: 1.5–98.2%). Publication bias was only found in eight studies (Egger's test: p < 0.05). Conclusions Our meta-analyses revealed important risk factors that are associated with severity and mortality of COVID-19.

Care during the COVID-19 pandemic hinges upon the existence of fast, safe, and highly sensitive diagnostic tools. Considering significant practical advantages of lung ultrasound (LUS) over other imaging techniques, but difficulties for doctors in pattern recognition, we aim to leverage machine learning toward guiding diagnosis from LUS. We release the largest publicly available LUS dataset for COVID-19 consisting of 202 videos from four classes (COVID-19, bacterial pneumonia, non-COVID-19 viral pneumonia and healthy controls). On this dataset, we perform an in-depth study of the value of deep learning methods for the differential diagnosis of lung pathologies. We propose a frame-based model that correctly distinguishes COVID-19 LUS videos from healthy and bacterial pneumonia data with a sensitivity of 0.90±0.08 and a specificity of 0.96±0.04. To investigate the utility of the proposed method, we employ interpretability methods for the spatio-temporal localization of pulmonary biomarkers, which are deemed useful for human-in-the-loop scenarios in a blinded study with medical experts. Aiming for robustness, we perform uncertainty estimation and demonstrate the model to recognize low-confidence situations which also improves performance. Lastly, we validated our model on an independent test dataset and report promising performance (sensitivity 0.806, specificity 0.962). The provided dataset facilitates the validation of related methodology in the community and the proposed framework might aid the development of a fast, accessible screening method for pulmonary diseases. Dataset and all code are publicly available at: https://github.com/BorgwardtLab/covid19_ultrasound.

Weis (1) (2)*, Aline Cuénod (3) (4), Bastian Rieck (1) (2), Felipe Llinares-López (1) (2), Olivier Dubuis (8), Susanne Graf (7), Claudia Lang (8), Michael Oberle (9), Maximilian Brackmann (10), Kirstine K. Søgaard (3) (4), Michael Osthoff (5) (6), Karsten Borgwardt (1) (2)*+, Adrian Egli (3) (4)*+