Characterization and linking of imaging and molecular patterns in hepatocellular carcinoma
Abstract/Contents
- Abstract
- A wealth of cancer imaging data is becoming available due to routine clinical care. Similarly, public research efforts are unprecedently making large genomic cancer data sets more accessible. These data present unique opportunities: with the right statistical tools we can improve our clinical decision making and biological knowledge of disease. One instance of disease where computational models can be used to improve medical treatment is hepatocellular carcinoma (HCC). HCC is the third most common cause of cancer-related death in men, accounting for three billion dollars in direct and indirect costs. In this dissertation, we present the application and development of methods that leverage clinical radiological image data and genomic data of cancer with a focus on hepatocellular carcinoma as an example. This work comprises applications of statistical tools and machine learning algorithms to create surrogate cancer biomarkers, and the development of statistical methods that improve clinical diagnosis and discovery of cancer mechanisms. First, (I) we evaluate inter-reader reliability in annotating semantic features on preoperative computed tomography and investigate the potential of semantic features of radiological images to provide a surrogate imaging biomarker to microvascular invasion. We then show the overall superiority of consensus models over single-reader semantic-based models. Second, (II) we employ a radiomic-based approach to develop an imaging signature and develop a radomics model with superior performance over previously reported semantic signatures. Third, (III) we develop a new model for gene regulatory inference and learning. Specifically, our framework is the first to combine Bayesian and module-based approaches. It tackles an important limitation in module-based methods by allowing probabilistic assignments of target genes to modules. Our method outperforms current methods on sparsity metrics, with similar performance in modeling module expression. We show that our method can recover a range of diverse liver functions from metabolic and immune to protein synthesis.
Description
Type of resource | text |
---|---|
Form | electronic resource; remote; computer; online resource |
Extent | 1 online resource. |
Place | California |
Place | [Stanford, California] |
Publisher | [Stanford University] |
Copyright date | 2021; ©2021 |
Publication date | 2021; 2021 |
Issuance | monographic |
Language | English |
Creators/Contributors
Author | Bakr, Shaimaa Hesham Ahmed |
---|---|
Degree supervisor | Napel, Sandy |
Thesis advisor | Napel, Sandy |
Thesis advisor | Gentles, Andrew J |
Thesis advisor | Gevaert, Olivier Michel Simonne |
Thesis advisor | Nishimura, Dwight George |
Thesis advisor | Pauly, John (John M.) |
Degree committee member | Gentles, Andrew J |
Degree committee member | Gevaert, Olivier Michel Simonne |
Degree committee member | Nishimura, Dwight George |
Degree committee member | Pauly, John (John M.) |
Associated with | Stanford University, Department of Electrical Engineering |
Subjects
Genre | Theses |
---|---|
Genre | Text |
Bibliographic information
Statement of responsibility | Shaimaa Bakr. |
---|---|
Note | Submitted to the Department of Electrical Engineering. |
Thesis | Thesis Ph.D. Stanford University 2021. |
Location | https://purl.stanford.edu/jy543qp2770 |
Access conditions
- Copyright
- © 2021 by Shaimaa Hesham Ahmed Bakr
- 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...