Data-Driven Computational Approaches in Pain Medicine
Public DepositedChronic pain affects a large percentage of the adult population in the United States. The goal of this project was to analyze data from a study about the effects of acupuncture treatment on veterans of the Persian Gulf War (1990-91) afflicted with what is known as Gulf War Illness (GWI) to determine for whom acupuncture effectively reduces chronic pain. A series of machine learning models were developed to gain insight into the key factors that help predict whether a patient’s chronic pain improved. Logistic Regression yielded the most accurate predictive models for pain improvement in patients. Drawing from the calculated feature importance and Logistic Regression modeling, the most important factors for the prediction are derived from the McGill Pain Scale, SF-36 questionnaire, Locus of Control questionnaire, Pittsburgh Sleep Quality Index, and Carroll Depression Scale.
- This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.
- Creator
- Subject
- Publisher
- Identifier
- 108741
- E-project-050323-212754
- Keyword
- Advisor
- Year
- 2023
- UN Sustainable Development Goals
- Date created
- 2023-05-03
- Resource type
- Major
- Source
- E-project-050323-212754
- Rights statement
- Last modified
- 2023-06-13
Relations
- In Collection:
Items
Permanent link to this page: https://digital.wpi.edu/show/zk51vm21k