Type
Text
Type
Dissertation
Advisor
Advisors: Huang, Chuan; Vaska, Paul; Jia, Shu; Zhu, Wei
Date
2017-12-01
Keywords
Biomedical engineering
Department
Department of Biomedical Engineering | Dissertation
Language
en
Source
This work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree.
Identifier
http://hdl.handle.net/11401/78320
Publisher
The Graduate School, Stony Brook University: Stony Brook, NY.
Format
application/pdf
Abstract
The dissertation focuses on the histopathology image analysis for cancer diagnosis and prognosis, based on the innovation of histopathological machine vision techniques. Histopathology images are regarded as the reference standard to identify diseases, and especially as a gold standard on cancer diagnosis. With the recent advance of the electronic scanners, digitized whole slide images (WSI) make it possible to analyze cancer tissues in high resolution and large-scale manner. At the same time, computer aided diagnosis (CAD) algorithms are being developed to detect cancer automatically, both in radiological and pathological field. However, those CAD algorithms are based on object segmentation and handcrafted features, which are not fully automatic. I developed innovative methods and frameworks to assist cancer diagnosis and prognosis automatically, without sophisticated feature extraction. The major projects are intermediate prostate cancer grading, cell nuclei segmentation using deep learning and nuclei segmentation evaluation through image synthesis. My research novelties include multi-resolution histopathology image analysis, fully automatic gland cancerous degree classification, and nuclei segmentation and synthesis using deep learning. | 128 pages
Recommended Citation
Zhou, Naiyun, "Cancer Diagnosis and Prognosis with Histopathology Image Analysis and Pattern Recognition" (2017). Stony Brook Theses and Dissertations Collection, 2006-2020 (closed to submissions). 3814.
https://commons.library.stonybrook.edu/stony-brook-theses-and-dissertations-collection/3814