Title: Integrating Geolocation and Genetic Data to Improve Opioid Use Disorder Risk Modelling
Student: Sybille Légitime
Major Advisor: Derek Aguiar
Associate Advisors: Dipak Dey, Bing Wang
Date/Time: Tuesday, July 12, 2022, 2:00pm EST
Location: HBL 1102 Conference Room/WebEx
https://uconn-cmr.webex.com/uconn-cmr/j.php?MTID=m719cdba3e21c1194050e743d323e9893
Meeting number: 2623 399 0221
Meeting password: xuPMSAdg252
Abstract: Opioid use disorder (OUD) is a chronic disorder characterized by the problematic use of opioids causing clinical impairment. The impact of OUD is growing in the United States, where in 2020, an estimated 2.7 million people suffered from OUD (up from 1.6 million in 2019) and nearly 69,000 people lost their lives. In response to this crisis, physicians have searched for ways to improve patient outcomes. The estimation of patient risk to develop OUD is one strategy that can inform alternative analgesic prescriptions and improve monitoring programs or intervention strategies. At present, risk estimation is typically based on questionnaires and self-reported data, and thus there is a need for an objective and comprehensive approach to evaluate susceptibility to OUD before the initial prescription. We develop a novel strategy that integrates genetic and mobility trace data for a more accurate estimation of OUD risk. Specifically, we build a suite of machine learning methods to predict susceptibility to OUD based on both single nucleotide polymorphisms and geolocation data. Importantly, our methods capture both the genetic and environmental (behavioral) variability of OUD risk. We develop a framework that synthesizes samples from experimental genetic and mobility trace data based on relative risk and comorbidity, and compare our approach with several methods that rely solely on genetic or environmental data. Our combined approach shows significant improvement over the baseline methods with respect to the area under the receiver operating characteristic and precision-recall curves, suggesting that our novel strategy can be used to improve OUD risk estimation and, ultimately, patient outcomes.
Join from a video or application
Dial 26233990221@uconn-cmr.webex.com
You can also dial 173.243.2.68 and enter your meeting number.
Join by phone
+1-415-655-0002 US Toll
Access code: 26233990221