American Cancer Society estimates that in 2021 nearly 300,000 women in the United States will be diagnosed with invasive breast cancer, and about 43,600 women will die from breast cancer. While many have access to health care and cancer screening, women from rural or underdeveloped communities often have limited access. Therefore, there is a need for an inexpensive and easy-to-use breast cancer identification device, which can be employed in small clinics to provide support to primary care physicians. This work aims to develop a method to characterize breast tumors and tissue using non-invasive imaging modalities. The proposed bimodal imaging system has tactile and multispectral imaging capabilities. Tactile imaging modality characterizes tumors by esti-mating their depth, size, and stiffness, along with the Tactile Index. Multispectral imaging modality identifies breast asymmetry, texture, and inflammation changes, together with the Spectral Index. These indices are combined with the BCRAT Index, the risk score devel¬oped by the National Institute of Health, to form the Multimodal Index for personalized breast cancer risk assessment.
In this study, we will describe the development of the bimodal imaging system. We will present the algorithms for tactile and multispectral modalities. Tactile and Multispec¬tral Profile Diagrams are developed to capture broad imaging signals in a compact and application-specific way. A Tactile Profile Diagram is a pictorial representation of the rel¬ative depth, size, and stiffness of the imaged tumor. A Multispectral Profile Diagram is a representative pattern image for breast tissue superficial optical properties. To classify the profile diagrams, we employ the Convolutional Neural Network deep learning method. We will describe the results of the experiments conducted using tissue-mimicking phan¬toms and human in-vivo experiments. The results demonstrate the ability of the method to classify and quantify tumor and tissue characteristics. Finally, we describe the method to calculate Multimodal Index for the malignancy risk assessment via tactile and multispectral imaging modalities and the risk probability based on the health records. / Electrical and Computer Engineering
| Identifer | oai:union.ndltd.org:TEMPLE/oai:scholarshare.temple.edu:20.500.12613/6859 |
| Date | January 2021 |
| Creators | Oleksyuk, Vira, 0000-0002-5071-2298 |
| Contributors | Won, Chang-Hee, 1967-, Picone, Joseph, Obeid, Iyad, 1975-, Pleshko, Nancy, Du, Xiaojiang |
| Publisher | Temple University. Libraries |
| Source Sets | Temple University |
| Language | English |
| Detected Language | English |
| Type | Thesis/Dissertation, Text |
| Format | 230 pages |
| Rights | IN COPYRIGHT- This Rights Statement can be used for an Item that is in copyright. Using this statement implies that the organization making this Item available has determined that the Item is in copyright and either is the rights-holder, has obtained permission from the rights-holder(s) to make their Work(s) available, or makes the Item available under an exception or limitation to copyright (including Fair Use) that entitles it to make the Item available., http://rightsstatements.org/vocab/InC/1.0/ |
| Relation | http://dx.doi.org/10.34944/dspace/6841, Theses and Dissertations |
Page generated in 0.1174 seconds