Date of Award

Fall 7-31-2022

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

School

Computing Sciences and Computer Engineering

Committee Chair

Dr. Beddhu Murali, Committee Chair

Committee Chair School

Computing Sciences and Computer Engineering

Committee Member 2

Dr. Andrew Sung

Committee Member 2 School

Computing Sciences and Computer Engineering

Committee Member 3

Dr. Sarah B. Lee

Committee Member 3 School

Computing Sciences and Computer Engineering

Committee Member 4

Dr. Amer Dawoud

Committee Member 4 School

Computing Sciences and Computer Engineering

Committee Member 5

Dr. Sungwook Lee

Committee Member 5 School

Mathematics and Natural Sciences

Abstract

Machine Learning and Artificial Intelligence have made significant progress concurrent with new advancements in hardware and software technologies. Deep learning methods heavily utilize parallel computing and Graphical Processing Units(GPU). It is already used in many applications ranging from image classification, object detection, segmentation, cyber security problems and others. Deep Learning is emerging as a viable choice in dealing with today’s real-time medical problems. We need new methods and technologies in the field of Medical Science and Epidemiology for detecting and diagnosing emerging threats from new viruses such as COVID-19. The use of Artificial Intelligence in these domains is becoming more accepted, not as a replacement for current medical practices but rather as an enhancing and augmenting tool to current practices. This dissertation is about the Application of Deep Learning for Medical Sciences and Epidemiology Data analysis and Diagnosis Modeling. The work started with an analysis of existing data, then focused on developing diagnosis models using both supervised and unsupervised approaches. This resulted in a renal transplantation recommender system, breast cancer diagnosis model and COVID-19 diagnosis models. The recent completed work for COVID-19 diagnosis modeling leverage the potential of more advanced deep neural network model and vision transformer. The results are very promising and outperformed all existing models. The final diagnosis model has three kinds of dataset. For Chest X-Ray(CXR) image dataset training accuracy is 95.9376 %, validation accuracy is 96.1667 % and testing accuracy is 95.1250 %. For Chest CT image dataset training accuracy is 96.5474 %, validation accuracy is 95.7302 % and testing accuracy is 97.0588 %. For the Combined CXR and Chest CT image dataset training accuracy is 95.6665 %, validation accuracy is 96.6961% and testing accuracy is 97.8859 %.

ORCID ID

https://orcid.org/ 0000-0002-3880-0308

Share

COinS