Mohammad Abbadi's research while affiliated with George Washington University and other places

Since the spread of the COVID-19 pandemic, the number of patients has increased dramatically, making it difficult for medical staff, including doctors, to cover hospitals and monitor patients. Therefore, this work depends on Computerized Tomography (CT) scan images to diagnose COVID-19. CT scan images are used to diagnose and determine the severity of the disease. On the other hand, Deep Learning (DL) is widely used in medical research, making great progress in medical technologies. For the diagnosis process, the Convolutional Neural Network (CNN) algorithm is used as a type of DL algorithm. Hence, this work focuses on detecting COVID-19 from CT scan images and determining the severity of the illness. The proposed model is as follows: first, classifying CT scan images into infected or not infected using one of the CNN structures, Residual Neural Networks (ResNet50); second, applying a segmentation process for the infected images to identify lungs and pneumonia using the SegNet algorithm (a CNN architecture for semantic pixel-wise segmentation) so that the disease's severity can be determined; finally, applying linear regression to predict the disease's severity for any new image. The proposed approach reached an accuracy of 95.7% in the classification process and lung and pneumonia segmentation of 98.6% and 96.2%, respectively. Furthermore, a regression process reached an accuracy of 98.29%.

The human hand has been considered a promising component for biometric-based identification and authentication systems for many decades. In this paper, hand side recognition framework is proposed based on deep learning and biometric authentication using the hashing method. The proposed approach performs in three phases: (a) hand image segmentation and enhancement by morphological filtering, automatic thresholding, and active contour deformation, (b) hand side recognition based on deep Convolutional Neural Networks (CNN), and (c) biometric authentication based on the hashing method. The proposed framework is evaluated using a very large hand dataset, which consists of 11076 hand images, including left/ right and dorsal/ palm hand images for 190 persons. Finally, the experimental results show the efficiency of the proposed framework in both dorsal-palm and left-right recognition with an average accuracy of 96.24 and 98.26, respectively, using a completely automated computer program.