ABSTRACT
Precise and fast diagnosis of COVID-19 cases play a vital role in early stage of medical treatment and prevention. Automatic detection of COVID-19 cases using the chest X-ray images and chest CT-scan images will be helpful to reduce the impact of this pandemic on the human society. We have developed a novel FractalCovNet architecture using Fractal blocks and U-Net for segmentation of chest CT-scan images to localize the lesion region. The same FractalCovNet architecture is also used for classification of chest X-ray images using transfer learning. We have compared the segmentation results using various model such as U-Net, DenseUNet, Segnet, ResnetUNet, and FCN. We have also compared the classification results with various models like ResNet5-, Xception, InceptionResNetV2, VGG-16 and DenseNet architectures. The proposed FractalCovNet model is able to predict the COVID-19 lesion with high F-measure and precision values compared to the other state-of-the-art methods. Thus the proposed model can accurately predict the COVID-19 cases and discover lesion regions in chest CT without the manual annotations of lesions for every suspected individual. An easily-trained and high-performance deep learning model provides a fast way to identify COVID-19 patients, which is beneficial to control the outbreak of SARS-II-COV.
ABSTRACT
BACKGROUND: Presymptomatic COVID-19 patients have been identified as a major stumbling block in efforts to break the chain of transmission. Studies on temporal dynamics of its shedding suggests it peaks 1-2 days prior to any symptom onset. Therefore, a large proportion of patients are actively spreading the disease unknowingly whilst undetected. However, lengthy lockdowns and isolation leads to a host of socioeconomic issues and are impractical. Conversely, there exists no study describing this group and their clinical significance despite their key role in disease transmission. METHODS: As a result, we devised a retrospective study to look at the prevalence of presymptomatic patients with COVID-19 from data sourced via our medical records office. Subsequently, we identify early indicators of infection through demographic information, biochemical and radiological abnormalities which would allow early diagnosis and isolation. In addition, we will look into the clinical significance of this group and their outcome; if it differs from asymptomatic or symptomatic patients. Descriptive statistics were used in addition to tabulating the variables and corresponding values for reference. Variables are compared between the presymptomatic group and others via Chi-square testing and Fisher's exact test, accepting a p value of < 0.05 as significant. RESULTS: Our analysis shows a higher proportion of presymptomatic patients with atypical symptoms like chest pain while symptomatic patients commonly present with respiratory symptoms like cough and shortness of breath. Besides that, there were more females presenting as presymptomatic patients compared to males (p = 0.019) and these group of patients were likely to receive treatment (p < 0.001). Otherwise, we were not able to identify other statistically significant markers suggesting a patient is presymptomatic. CONCLUSION: As we have little means of identifying these silent spreaders, it highlights further the importance of general measures implemented to stop COVID-19 transmission like social distancing, face mask, and widespread testing.
