Development and validation of chest CT-based imaging biomarkers for early stage COVID-19 screening

Coronavirus Disease 2019 (COVID-19) is currently a global pandemic, and early screening is one of the key factors for COVID-19 control and treatment. Here, we developed and validated chest CT-based imaging biomarkers for COVID-19 patient screening from two independent hospitals with 419 patients. We identified the vasculature-like signals from CT images and found that, compared to healthy and community acquired pneumonia (CAP) patients, COVID-19 patients display a significantly higher abundance of these signals. Furthermore, unsupervised feature learning led to the discovery of clinical-relevant imaging biomarkers from the vasculature-like signals for accurate and sensitive COVID-19 screening that have been double-blindly validated in an independent hospital (sensitivity: 0.941, specificity: 0.920, AUC: 0.971, accuracy 0.931, F1 score: 0.929). Our findings could open a new avenue to assist screening of COVID-19 patients.

Effect of synchronous online vs. face-to-face cardiopulmonary resuscitation training on chest compression quality: A pilot randomized manikin study.

OBJECTIVES: The aim of the study was to compare the effect of synchronous online and face-to-face cardiopulmonary resuscitation (CPR) training on chest compressions quality in a manikin model. METHODS: A total of 118 fourth-year medical students participated in this study. The participants were divided into two groups: the online synchronous teaching group and the face-to-face group. Then, the participants were further randomly distributed to 1 of 2 feedback groups: online synchronous teaching and training with feedback devices (TF, n = 30) or without feedback devices (TN, n = 29) and face-to-face teaching and training with feedback devices (FF, n = 30) or without feedback devices (FN, n = 29). In the FN group and FF group, instructors delivered a 45-min CPR training program and gave feedback and guidance during training on site. In the TN group and TF group, the participants were trained with an online lecture via Tencent Meeting live broadcasting. Finally, participants performed a 2-min continuous chest compression (CC) during a simulated cardiopulmonary arrest scene without the audiovisual feedback (AVF) device. The outcome measures included CC depth, CC rate, proportions of appropriate depth (50-60 mm) and CC rate (100-120/min), percentage of correct hand location position, and percentage of complete chest recoil. RESULTS: There was little difference in the CC quality between the synchronous online training groups and the face-to-face training groups. There was no statistically significant difference in CC quality between the TN group and FN group. There were also no statistically significant differences between the TF and FF groups in terms of correct hand position, CC depth, appropriate CC depth, complete chest recoil or CC rate. However, the FF group had a higher appropriate CC rate than the TF group (p = 0.045). In the face-to-face training groups, the AVF device group had a significantly greater CC depth, appropriate CC depth, CC rate, and appropriate CC rate. However, there was a lack of statistically significant differences in terms of correct hand position (p = 0.191) and appropriate CC depth (p = 0.123). In the synchronous online training groups, the AVF device had little effect on the CC rate (p = 0.851) and increased the appropriate CC rate, but the difference was not statistically significant (p = 0.178). CONCLUSIONS: Synchronous online training with an AVF device would be a potential alternative approach to face-to-face chest compression training. Synchronous online training with AVF devices seems to be a suitable replacement for face-to-face training to offer adequate bystander CPR chest compression training.

Wearing a N95 mask increases rescuer's fatigue and decreases chest compression quality in simulated cardiopulmonary resuscitation.

OBJECTIVES: N95 mask is essential for healthcare workers dealing with the coronavirus disease 2019 (COVID-19). However, N95 mask causes discomfort breathing with marked reduction in air exchange. This study was designed to investigate whether the use of N95 mask affects rescuer's fatigue and chest compression quality during cardiopulmonary resuscitation (CPR). METHODS: After a brief review of CPR, each participant performed a 2-minute continuous chest compression on a manikin wearing N95 (N95 group, n = 40) or surgical mask (SM group, n = 40). Compression rate and depth, the proportions of correct compression rate, depth, complete chest recoil and hand position were documented. Participants' fatigue was assessed using Borg score. RESULTS: Significantly lower mean chest compression rate and depth were both achieved in the N95 group than in the SM group (p < 0.05, respectively). In addition, the proportion of correct compression rate (61 ± 19 vs. 75 ± 195, p = 0.0067), depth (67 ± 16 vs. 90 ± 14, p < 0.0001) and complete recoil (91 ± 16 vs. 98 ± 5%, p = 0.0248) were significantly decreased in the N95 group as compared to the SM group. At the end of compression, the Borg score in the N95 group was significantly higher than that in the SM group (p = 0.027). CONCLUSION: Wearing a N95 mask increases rescuer's fatigue and decreases chest compression quality during CPR. Therefore, the exchange of rescuers during CPR should be more frequent than that recommended in current guidelines when N95 masks are applied.