Affordable measures to monitor and alarm nosocomial SARS-CoV-2 infection due to poor ventilation.

Since the coronavirus disease 2019 (COVID-19) outbreak, the nosocomial infection rate worldwide has been reported high. It is urgent to figure out an affordable way to monitor and alarm nosocomial infection. Carbon dioxide (CO2 ) concentration can reflect the ventilation performance and crowdedness, so CO2 sensors were placed in Beijing Tsinghua Changgung Hospital's fever clinic and emergency department where the nosocomial infection risk was high. Patients' medical records were extracted to figure out their timelines and whereabouts. Based on these, site-specific CO2 concentration thresholds were calculated by the dilution equation and sites' risk ratios were determined to evaluate ventilation performance. CO2 concentration successfully revealed that the expiratory tracer was poorly diluted in the mechanically ventilated inner spaces, compared to naturally ventilated outer spaces, among all of the monitoring sites that COVID-19 patients visited. Sufficient ventilation, personal protection, and disinfection measures led to no nosocomial infection in this hospital. The actual outdoor airflow rate per person (Qc ) during the COVID-19 patients' presence was estimated for reference using equilibrium analysis. During the stay of single COVID-19 patient wearing a mask, the minimum Qc value was 15-18 L/(s·person). When the patient was given throat swab sampling, the minimum Qc value was 21 L/(s·person). The Qc value reached 36-42 L/(s·person) thanks to window-inducted natural ventilation, when two COVID-19 patients wearing masks shared the same space with other patients or healthcare workers. The CO2 concentration monitoring system proved to be effective in assessing nosocomial infection risk by reflecting real-time dilution of patients' exhalation.

AI-assisted CT imaging analysis for COVID-19 screening: Building and deploying a medical AI system.

The sudden outbreak of novel coronavirus 2019 (COVID-19) increased the diagnostic burden of radiologists. In the time of an epidemic crisis, we hope artificial intelligence (AI) to reduce physician workload in regions with the outbreak, and improve the diagnosis accuracy for physicians before they could acquire enough experience with the new disease. In this paper, we present our experience in building and deploying an AI system that automatically analyzes CT images and provides the probability of infection to rapidly detect COVID-19 pneumonia. The proposed system which consists of classification and segmentation will save about 30%-40% of the detection time for physicians and promote the performance of COVID-19 detection. Specifically, working in an interdisciplinary team of over 30 people with medical and/or AI background, geographically distributed in Beijing and Wuhan, we are able to overcome a series of challenges (e.g. data discrepancy, testing time-effectiveness of model, data security, etc.) in this particular situation and deploy the system in four weeks. In addition, since the proposed AI system provides the priority of each CT image with probability of infection, the physicians can confirm and segregate the infected patients in time. Using 1,136 training cases (723 positives for COVID-19) from five hospitals, we are able to achieve a sensitivity of 0.974 and specificity of 0.922 on the test dataset, which included a variety of pulmonary diseases.