ABSTRACT
Radiography is used in medical treatment as a method to diagnose the internal organs of the human body from diseases. However, the advancement in machine learning technologies have paved way to new possibilities of diagnosing diseases from chest X-ray images. One such diseases that are able to be detected by using X-ray is the COVID-19 coronavirus. This research investigates the diagnosis of COVID-19 through X-ray images by using transfer learning and fine-tuning of the fully connected layer. Hyperparameters such as dropout, p, number of neurons, and activation functions are investigated on which combinations of these hyperparameters will yield the highest classification accuracy model. VGG19 learning model created by the Visual Geometry Group is used for extraction of features from the patient's chest X-ray images. To evaluate the combination of various pipelines, the loss and accuracy graphs are used to find the pipeline which performs the best in classification task. The findings in this research will open new possibilities in screening method for COVID-19.
ABSTRACT
Transfer Learning (TL) opens new possibilities of detection of disease through radiography as compared to conventional machine learning as well as deep learning methods. The extraction of features through pre-trained Convolutional Neural Networks (CNN) and the tuning of the fully connected layers of the CNN model is the core for the development of a transfer learning pipeline. The present study investigates the diagnosis of COVID-19 through X-ray images by means of three TL models, namely Inception V3, VGG-16, and the VGG-19 for feature extraction along with heuristically fine-tuned fully connected layers. It was demonstrated through this preliminary work that both the VGG-16 and VGG-19 tuned pipelines could achieve a train and test classification accuracies of 99.8% and 94%, respectively. © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.