Liora Mayats Alpay
Major/Minor: Computer Science
Title: Machine Learning enhanced COVID-19 detection for classifying chest X-ray images.
Abstract: Due to the radiographic similarities between pneumonia resulting from COVID-19 infection and that caused by other infections, it is challenging and time-consuming for radiologists to diagnosis COVID-19. Developing a deep learning model for automatic COVID-19 diagnosis from chest X-Ray images can significantly help address the challenges of the present pandemic.
In this project, we develop Convolutional Neural Network models based on VGG-16 and ResNet50 deep learning framework. We implement automatic detection of COVID-19, assessing the presence of COVID-19 on chest radiography, and classifying between Healthy patients, Pneumonia patients, and COVID-19 patients. The approaches are validated using 3000 publicly accessible chest X-ray images. Moreover, to demonstrate the framework's accuracy, the Gradient-weighted Class Activation Mapping (Grad-Cam) is adapted to visualize and explain the predicted results. The approach is generic and can be applied to other medical classification and detection problems, such as classifying and detecting different types of cancer.
