Leveraging Deep Learning and Radiography for Diagnosis Of COVID-19
Today, Machine learning and Deep learningare being leveraged by almost every industry all over the world. They have now become ubiquitous tools for research in the fields of Robotics, Autonomous cars, Computer vision, medical and health sciences to name a few. In this article the equip frontiers of Deep learning and Computer Vision are used to diagnose COVID-19. The primary clinical method that is currently in use for the diagnosis of COVID-19, Reverse Transcription Polymerase Chain Reaction (RTPCR) is expensive and requires trained medical personnel. Radiography is an easily accessible tool that can be a reasonable alternative to RT-PCR in diagnosing COVID-19. A Convolutional Neural Network based on VGG16 architecture is trained and analyzed on around 21000 lung X-ray images using transfer learning. Out of the 21170 images obtained from kaggle repository, 16500 images have been used for training, 3130 have been used for validation and 1540 for testing the validated model. The goal is to accurately screen the patients suffering from Covid-19 against those who also suffer from Ground Glass Opacity and Viral Pneumonia which have a similar effect on human lungs as that of Covid-19. In the result analysis, the model gives a train accuracy of 99% and a validation accuracy of 94.16%. The proposed model helps radiologists diagnose COVID-19 within 0.5 seconds in a system equipped with GPU (Graphic Processing Unit) by classifying thousands or even millions of images in a single click. When trained with a larger dataset, the model may lead to facilitating early treatment of such lethal disease resulting in improved clinical outcomes. This work proposes a possible method of screening COVID-19 infected patientsbut do not claim any medical accuracy.
Keywords-COVID-19, VGG16, TPU, Ground Glass Opacity, Viral Pneumonia, Scatter plot, Sofmax, ReLU, Keras, Categorical Cross Entropy, Adam Optimizer, Precision, Accuracy, Recall, F1 score, Confusion matrix.