Image processing in pathology for the discovery of clinically relevant disease subtypes
Abstract/Contents
- Abstract
- Computational analysis of images in medicine is unlocking new information in medical data, allowing for advances in health care. While this field has matured to the point of clinical use in radiology, its applications in pathology are younger, and more in need of development to reach its full potential. Here we present four advances in medical image analysis in pathology that have demonstrated potential to translate into better health care. First, we describe and evaluate two novel image pre-processing methods in the fields of 1) stain normalization and 2) nuclear segmentation. Both methods show an improvement over established methods in the literature, not only in direct evaluation, but also in downstream applications. Next, we describe and rigorously analyze the stability of an image processing pipeline for translating quantitative pathology data into clinically meaningful information. Finally, we apply the previous methods to discover a novel finding: a link between pathology images and gender. The gender based differences we discovered in the pathology data can be used to create four distinct subtypes, each with a distinct overall survival profile. These differences in pathology correspond to newly discovered molecular differences where four genes are associated with a beneficial prognosis in males and a detrimental prognosis in females. The gene expression differences indicates a putative mechanism for the survival differences, indicating that the techniques outlined here can be used not only for clinical applications, but for basic biological research as well.
Description
Type of resource | text |
---|---|
Form | electronic; electronic resource; remote |
Extent | 1 online resource. |
Publication date | 2015 |
Issuance | monographic |
Language | English |
Creators/Contributors
Associated with | Barker, Jocelyn E |
---|---|
Associated with | Stanford University, Department of Biophysics. |
Primary advisor | Rao, Jianghong |
Primary advisor | Rubin, Daniel (Daniel L.) |
Thesis advisor | Rao, Jianghong |
Thesis advisor | Rubin, Daniel (Daniel L.) |
Thesis advisor | Napel, Sandy |
Thesis advisor | West, Robert |
Advisor | Napel, Sandy |
Advisor | West, Robert |
Subjects
Genre | Theses |
---|
Bibliographic information
Statement of responsibility | Jocelyn E. Barker. |
---|---|
Note | Submitted to the Department of Biophysics. |
Thesis | Thesis (Ph.D.)--Stanford University, 2015. |
Location | electronic resource |
Access conditions
- Copyright
- © 2015 by Jocelyn Elaine Barker
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...