2026-09-21T04:34:10Zhttps://dspace.mit.edu/server/oai/request

oai:dspace.mit.edu:1721.1/1072832026-06-17T14:44:06Zcom_1721.1_7582com_1721.1_7581col_1721.1_131022

Polina Golland. Dalca, Adrian Vasile Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science. Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science 2017-03-10T14:19:41Z 2017-03-10T14:19:41Z 2016 2016 http://hdl.handle.net/1721.1/107283 972902575 Thesis: Ph. D., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2016. This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections. Cataloged from student-submitted PDF version of thesis. Includes bibliographical references (pages 125-141). We develop mathematical models that exploit external information to improve analysis of a medical scan. Medical images enable visualization of the human body, and are central in clinical practice and many large-scale scientific studies. Medical image analysis uses computational models to interpret these scans towards the clinical or research goals. For example, in this thesis we are motivated by a clinical study of ischemic stroke, which aims to quantify cerebrovascular disease burden as observed in medical scans, along with its population trends and genetic predisposition. In most analyses, anatomical information is extracted from images to provide insight into a problem, facilitating understanding of genetic variants, clinical variables and population trends. In contrast, this thesis investigates what these external factors tell us about the human anatomy and the medical scans themselves. First, we show how genetic and clinical indicators can be used to predict MRI scans of anatomical change through a semi-parametric generative model. Second, we demonstrate that a cohort of subjects with cerebrovascular disease can help identify the spatially complex pathology in a new subject through a generative computational model. Third, we use large collections of clinical images to dramatically improve the resolution of a new scan and recover ne-scale anatomy. We also present an approach for rapid interactive visualization of images in large studies. Bringing our methods together in large scale analyses of stroke and dementia subjects, we demonstrate new avenues of research enabled by these contributions. by Adrian Vasile Dalca. Ph.D. 141 pages eng Massachusetts Institute of Technology MIT theses are protected by copyright. They may be viewed, downloaded, or printed from this source but further reproduction or distribution in any format is prohibited without written permission. http://dspace.mit.edu/handle/1721.1/7582 Electrical Engineering and Computer Science. Genetic, clinical and population priors for brain images Thesis application/pdf false Publication unknown unknown <Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="89331c37-2d95-4bb4-bd44-05e93b94739d"> <Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type> <Language>eng</Language> <Title>Genetic, clinical and population priors for brain images</Title> <PublishedIn> <Publication> </Publication> </PublishedIn> <PublicationDate>2016</PublicationDate> <Authors> <Author> <DisplayName>Dalca, Adrian Vasile</DisplayName> <Affiliation> <OrgUnit> </OrgUnit> </Affiliation> </Author> </Authors> <Editors> </Editors> <Publishers> <Publisher> <DisplayName>Massachusetts Institute of Technology</DisplayName> <OrgUnit /> </Publisher> </Publishers> <License>http://dspace.mit.edu/handle/1721.1/7582</License> <Keyword>Electrical Engineering and Computer Science.</Keyword> <Abstract>We develop mathematical models that exploit external information to improve analysis of a medical scan. Medical images enable visualization of the human body, and are central in clinical practice and many large-scale scientific studies. Medical image analysis uses computational models to interpret these scans towards the clinical or research goals. For example, in this thesis we are motivated by a clinical study of ischemic stroke, which aims to quantify cerebrovascular disease burden as observed in medical scans, along with its population trends and genetic predisposition. In most analyses, anatomical information is extracted from images to provide insight into a problem, facilitating understanding of genetic variants, clinical variables and population trends. In contrast, this thesis investigates what these external factors tell us about the human anatomy and the medical scans themselves. First, we show how genetic and clinical indicators can be used to predict MRI scans of anatomical change through a semi-parametric generative model. Second, we demonstrate that a cohort of subjects with cerebrovascular disease can help identify the spatially complex pathology in a new subject through a generative computational model. Third, we use large collections of clinical images to dramatically improve the resolution of a new scan and recover ne-scale anatomy. We also present an approach for rapid interactive visualization of images in large studies. Bringing our methods together in large scale analyses of stroke and dementia subjects, we demonstrate new avenues of research enabled by these contributions.</Abstract> <Access xmlns="http://purl.org/coar/access_right" > </Access> </Publication>