ABSTRACT
With the global outbreak of Corona Virus Disease 2019(COVID-19), many countries had made it mandatory for people to wear masks in public places. This paper proposed a novel mask detection algorithm RMPC (Restructing the Maxpool layer and the Convolution layer)-YOLOv7 based on YOLOv7 for detecting whether people wear masks in public places. The RMPC-YOLOv7 algorithm reconstructed the downsampling structure in the original YOLOv7 algorithm. We changed the stacking of the maxpooling layer and the convolutional layer. This enabled the feature information to be fully integrated to achieve the accuracy improvement of the new model. Through comparison experiments, our proposed RMPC-YOLOv7 had was improved 0.9% and 1.2% for mAP0.5 and mAP0.5:0.95, respectively. The experimental results demonstrated the feasibility of RMPC-YOLOv7. © 2023 IEEE.
ABSTRACT
Objectives: Prior studies have documented multiple organ injuries related to COVID-19. These findings raisedconcerns on the impact of injuries on the health of patients discharged. We aim to identify and evaluate the minimalresidual lesions as well as the critical organs by applying the 18F-FDG PET/CT dynamic modeling approach fordischarged COVID-19 patients. Methods: Seven healthy volunteers (without COVID-19 infection or tumor history) completed static 18F-FDGPET/CT scan as control group 1. Dynamic 18F-FDG PET scans were performed for 19 patients including 7discharged COVID-19 patients (COVID-19 group) and 12 non-metastatic, newly diagnosed thyroid cancer (controlgroup 2). Dynamic 18F-FDG PET/CT scans were carried out two to three months after discharge for COVID-19group. Dynamic scans were performed from the apex of lung to the lower margin of liver. Patlak graphic analysis (Ki calculation) based on the dynamic modeling and SUV analysis from conventional static data were performed. Inaddition, COVID-19 group also completed lung function test one month after discharge. Results: COVID-19 patients were consecutively recruited, consisting of four severe and three mild pneumoniacases. Figure 1A shows the time course of recovery for a representative COVID-19 patient. One patient suffered anincreased 18F-FDG uptake of the lower lobe of bilateral lungs and a slightly impaired lung function. Although nosignificant differences were found for SUV between COVD-19 group and control group 1 in myocardial ( p =0.1677) and liver ( p = 0.0701), the SUV of lungs, brain, spleen, kidney, cervical vertebrae, thoracic vertebrae, lumbar vertebrae, and pelvisacrum in the COVID-19 group were significantly higher than that in the control group 1( p < 0.05 for all;Figure 1B ). Consistent with static results, the Ki values of myocardial and liver were notsignificantly different between COVID-19 group and control group 2. However, the Ki values of lung from severepneumonia were significantly higher than that of mild pneumonia ( p = 0.0034) and control group 2 ( p = 0.0003;Figure 1C ). Conclusion: Given such small number of enrolled, this preliminary study suggested that 18F-FDGPET/CT provided a reliable estimate of residual lesions and may act as a potential tool for evaluating recoverystatus in discharged COVID-19 subjects.