Utilization of images in an open source software to detect COVID-19
-
Abstract
COVID-19 pandemic has affected the lives of many people across the world. The symptoms of COVID-19 range from being mild to severe causing the patients to suffer from breathlessness, cough, cold, etc. Patients having severe symptoms also have side effects even after they have recovered from it and it may take even months to overcome those side effects. COVID-19 mostly affects our lungs with adverse infection due to which a person is unable to breathe with decorum. This research chapter employs models to differentiate between COVID-19 fever with non-COVID-19 pneumonia from the CT scan images of lungs. The images have been taken from an open source software which consists of 1,252 CT scan images of SARS-CoV-2 positive patients and 1,229 CT scan images of SARS-CoV-2 negative patients. The authors have used six models of Convolutional Neural Network (CNN) and Support Vector Machine (SVM) for detection of COVID-19. The authors conducted comparison between the models of CNN and SVM using different performance measures. The authors observed superior performance of CNN as compared to SVM. The results have been promising which encourages more prediction and classification using deep learning in other fields of medical sciences as well.
Abstract
COVID-19 pandemic has affected the lives of many people across the world. The symptoms of COVID-19 range from being mild to severe causing the patients to suffer from breathlessness, cough, cold, etc. Patients having severe symptoms also have side effects even after they have recovered from it and it may take even months to overcome those side effects. COVID-19 mostly affects our lungs with adverse infection due to which a person is unable to breathe with decorum. This research chapter employs models to differentiate between COVID-19 fever with non-COVID-19 pneumonia from the CT scan images of lungs. The images have been taken from an open source software which consists of 1,252 CT scan images of SARS-CoV-2 positive patients and 1,229 CT scan images of SARS-CoV-2 negative patients. The authors have used six models of Convolutional Neural Network (CNN) and Support Vector Machine (SVM) for detection of COVID-19. The authors conducted comparison between the models of CNN and SVM using different performance measures. The authors observed superior performance of CNN as compared to SVM. The results have been promising which encourages more prediction and classification using deep learning in other fields of medical sciences as well.
Chapters in this book
- Frontmatter I
- Preface V
- Acknowledgment VII
- Contents IX
- Editors’ profile XI
- Revolutionary transformations in twentieth century: making AI-assisted software development 1
- Useful techniques and applications of computational intelligence 19
- Machine learning-based attribute value search technique software component retrieval 33
- Use of fuzzy logic approach for software quality evaluation in Agile software development environment 55
- Metaheuristics for empirical software measurements 67
- Use of genetic algorithms in software testing models 81
- Insights into DevOps automation tools employed at different stages of software development 93
- Study of computational techniques to deal with ambiguity in SRS documents 107
- Utilization of images in an open source software to detect COVID-19 121
- Designing of a tool for comparing and analyzing different test suites of open source software 143
- Decision tree–based improved software fault prediction: a computational intelligence approach 163
- Performance analysis for SDN POX and open daylight controller using network emulator Mininet under DDoS attack 177
- Index 199
Chapters in this book
- Frontmatter I
- Preface V
- Acknowledgment VII
- Contents IX
- Editors’ profile XI
- Revolutionary transformations in twentieth century: making AI-assisted software development 1
- Useful techniques and applications of computational intelligence 19
- Machine learning-based attribute value search technique software component retrieval 33
- Use of fuzzy logic approach for software quality evaluation in Agile software development environment 55
- Metaheuristics for empirical software measurements 67
- Use of genetic algorithms in software testing models 81
- Insights into DevOps automation tools employed at different stages of software development 93
- Study of computational techniques to deal with ambiguity in SRS documents 107
- Utilization of images in an open source software to detect COVID-19 121
- Designing of a tool for comparing and analyzing different test suites of open source software 143
- Decision tree–based improved software fault prediction: a computational intelligence approach 163
- Performance analysis for SDN POX and open daylight controller using network emulator Mininet under DDoS attack 177
- Index 199
