Early detection of diseases such as COVID-19 could be a critical tool in reducing disease transmission by helping individuals recognize when they should self-isolate, seek testing, and obtain early medical intervention. Consumer wearable devices that continuously measure physiological metrics hold promise as tools for early illness detection. We gathered daily questionnaire data and physiological data using a consumer wearable (Oura Ring) from 63,153 participants, of whom 704 self-reported possible COVID-19 disease. We selected 73 of these 704 participants with reliable confirmation of COVID-19 by PCR testing and high-quality physiological data for algorithm training to identify onset of COVID-19 using machine learning classification. The algorithm identified COVID-19 an average of 2.75 days before participants sought diagnostic testing with a sensitivity of 82% and specificity of 63%. The receiving operating characteristic (ROC) area under the curve (AUC) was 0.819 (95% CI [0.809, 0.830]). Including continuous temperature yielded an AUC 4.9% higher than without this feature. For further validation, we obtained SARS CoV-2 antibody in a subset of participants and identified 10 additional participants who self-reported COVID-19 disease with antibody confirmation. The algorithm had an overall ROC AUC of 0.819 (95% CI [0.809, 0.830]), with a sensitivity of 90% and specificity of 80% in these additional participants. Finally, we observed substantial variation in accuracy based on age and biological sex. Findings highlight the importance of including temperature assessment, using continuous physiological features for alignment, and including diverse populations in algorithm development to optimize accuracy in COVID-19 detection from wearables.

Background and study aim Mastering laparoscopic surgical skills requires considerable time and effort. The Virtual Basic Laparoscopic Skill Trainer (VBLaST-PT©) is being developed as a computerized version of the peg transfer task of the Fundamentals of Laparoscopic Surgery (FLS) system using virtual reality technology. We assessed the learning curve of trainees on the VBLaST-PT© using the cumulative summation (CUSUM) method and compared them with those on the FLS to establish convergent validity for the VBLaST-PT©. Methods Eighteen medical students from were assigned randomly to one of three groups: control, VBLaST-training and FLS-training. The VBLaST and the FLS groups performed a total of 150 trials of the peg-transfer task over a three week period, five days a week. Their CUSUM scores were computed based on pre-defined performance criteria (junior, intermediate and senior levels). Results Of the six subjects in the VBLaST-training group, five achieved at least the “junior” level, three achieved the “intermediate” level and one achieved the “senior” level of performance criterion by the end of the 150 trials. In comparison, for the FLS group, three students achieved the “senior” criterion and all six students achieved the “intermediate” and “junior” criteria by the 150th trials. Both the VBLaST-PT© and the FLS systems showed significant skill improvement and retention, albeit with system specificity as measured by transfer of learning in the retention test: The VBLaST-trained group performed better on the VBLaST-PT© than on FLS (p=0.003), while the FLS-trained group performed better on the FLS than on VBLaST-PT© (p=0.002). Conclusion We characterized the learning curve for a virtual peg transfer task on the VBLaST-PT© and compared it with the FLS using CUSUM analysis. Subjects in both training groups showed significant improvement in skill performance, but the transfer of training between systems was not significant.

No abstract

No abstract