2021/05/20 by Prerak Mann, Mann, Prerak, Sahaj Jain +5
Medicine · Computer Science · #COVID-19 diagnosis using AI #Radiomics and Machine Learning in Medical Imaging #AI in cancer detection
paper · pdf · doi:10.48550/arxiv.2105.11241
SARS-CoV-2, also known as COVID-19 or Coronavirus, is a viral contagious\ndisease that is infected by a novel coronavirus, and has been rapidly spreading\nacross the globe. It is very important to test and isolate people to reduce\nspread, and from here comes the need to do this quickly and efficiently.\nAccording to some studies, Chest-CT outperforms RT-PCR lab testing, which is\nthe current standard, when diagnosing COVID-19 patients. Due to this, computer\nvision researchers have developed various deep learning systems that can\npredict COVID-19 using a Chest-CT scan correctly to a certain degree. The\naccuracy of these systems is limited since deep learning neural networks such\nas CNNs (Convolutional Neural Networks) need a significantly large quantity of\ndata for training in order to produce good quality results. Since the disease\nis relatively recent and more focus has been on CXR (Chest XRay) images, the\navailable chest CT Scan image dataset is much less. We propose a method, by\nutilizing GANs, to generate synthetic chest CT images of both positive and\nnegative COVID-19 patients. Using a pre-built predictive model, we concluded\nthat around 40% of the generated images are correctly predicted as COVID-19\npositive. The dataset thus generated can be used to train a CNN-based\nclassifier which can help determine COVID-19 in a patient with greater\naccuracy.\n