Author ORCID Identifier

Date of Award

Summer 8-11-2020

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Computer Science

First Advisor

Dr. Saeid Belkasim

Second Advisor

Dr. Yanqing Zhang

Third Advisor

Dr. Pavel Skums

Fourth Advisor

Dr. Marina Arav


The effective understanding of the biological behavior and prognosis of cancer subtypes is becoming very important in-patient administration. Cancer is a diverse disorder in which a significant medical progression and diagnosis for each subtype can be observed and characterized. Computer-aided diagnosis for early detection and diagnosis of many kinds of diseases has evolved in the last decade. In this research, we address challenges associated with multi-organ disease diagnosis and recommend numerous models for enhanced analysis. We concentrate on evaluating the Magnetic Resonance Imaging (MRI), Computed Tomography (CT), and Positron Emission Tomography (PET) for brain, lung, and breast scans to detect, segment, and classify types of cancer from biomedical images. Moreover, histopathological, and genomic classification of cancer prognosis has been considered for multi-organ disease diagnosis and biomarker recommendation. We considered multi-modal, multi-class classification during this study. We are proposing implementing deep learning techniques based on Convolutional Neural Network and Generative Adversarial Network.

In our proposed research we plan to demonstrate ways to increase the performance of the disease diagnosis by focusing on a combined diagnosis of histology, image processing, and genomics. It has been observed that the combination of medical imaging and gene expression can effectively handle the cancer detection situation with a higher diagnostic rate rather than considering the individual disease diagnosis. This research puts forward a blockchain-based system that facilitates interpretations and enhancements pertaining to automated biomedical systems. In this scheme, a secured sharing of the biomedical images and gene expression has been established. To maintain the secured sharing of the biomedical contents in a distributed system or among the hospitals, a blockchain-based algorithm is considered that generates a secure sequence to identity a hash key. This adaptive feature enables the algorithm to use multiple data types and combines various biomedical images and text records. All data related to patients, including identity, pathological records are encrypted using private key cryptography based on blockchain architecture to maintain data privacy and secure sharing of the biomedical contents.

File Upload Confirmation