Semantic driven data migration for predictive treatment of major depressive disorder

Ashby, Brendan Evans
Thumbnail Image
Other Contributors
Varela, Carlos A.
Luciano, Joanne S.
Krishnamoorthy, M. S.
Issue Date
Computer science
Terms of Use
Attribution-NonCommercial-NoDerivs 3.0 United States
This electronic version is a licensed copy owned by Rensselaer Polytechnic Institute, Troy, NY. Copyright of original work retained by author.
Full Citation
SEMMDD is a collection of researchers working to improve treatment for Major Depressive Disorder. Specifically, focus is put on accurately modeling patient response to various depression therapies. This paper contributes to this ongoing research through two main bodies of work. First, a process is described that migrates traditional tabular datasets into a decentralized web of knowledge known as the Semantic Web. Data migration is achieved by employing World Wide Web Consortium open standards and best practices. Datasets altered in this way benefit by increasing their exposure to like-minded content creators and lowering the learning curve for others to become involved. Additionally, I explain how this process fulfills the need for a data migration system that facilitates effective data proliferation along with exhaustive provenance capture. The second body of work tasked Machine Learning algorithms to make predictions for a patient's ultimate response to therapy based on their initial depressive state. As a result, I make recommendations concerning which algorithm configurations best fit the depression data and where future modeling efforts can best be focused. Machine Learning algorithms were utilized to give an alternate perspective to modeling patient treatment response in comparison to a historical use of neural network modeling. This recent body of work contributes to a larger initiative working to empower clinicians with the tools necessary to prescribe medical treatment. Armed with this knowledge, clinicians can make treatment decisions based more on historical trends seen in data and less on anecdotal experience.
December 2014
School of Science
Dept. of Computer Science
Rensselaer Polytechnic Institute, Troy, NY
Rensselaer Theses and Dissertations Online Collection
CC BY-NC-ND. Users may download and share copies with attribution in accordance with a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License. No commercial use or derivatives are permitted without the explicit approval of the author.