Affiliations: | DeBakey Executive Research Leadership Program |
Project Leader: | Irina Gaynanova, Ph.D. irinag@tamu.edu Statistics |
Meeting Times:
|
TBD, but most likely Tue, Thu or Fri (once per week) |
Team Size:
|
4
|
Open Spots: | 0 |
Special Opportunities:
|
– earn STAT 491 credit
– flexible schedule besides required meeting time
– remote work beyond meeting times (with possibility of hybrid mode)
– potential of co-authorship on developed software products and resulting publications
|
Team Needs:
|
Must: Knowledge and experience working with python programming language, no conflict with weekly meeting times once set, willingness to learn on the spot, willingness to read medical literature and literature on statistical/machine learning methods as needed. Big plus/priority will be given to candidates with: experience with git and GitHub , knowledge of R programming language, previous experience with PyTorch or TensorFlow, previous experience with diabetes research. |
Description:
|
Data from wearable devices, such as continuous glucose monitors (CGM), activity trackers, ambulatory blood pressure monitors, and sleep EEG monitors, are increasingly common. This wealth of data has the potential to improve health outcomes and improve the management of various chronic diseases, e.g. diabetes. The purpose of this project is to develop software and statistical/machine learning methods to facilitate analyses of such data in the context of diabetes research, as well as to perform explorative analysis of possible relationships between data from different wearable monitors. This is a long-term project, with the specific directions each semester driven by the previous findings and team developments. For Fall 2022, the focus will be on processing data from continuous glucose monitors, implementing state-of-the-art machine learning methods for glucose predictions, and comparing their performance. Some specific tasks that students will be expected to do as part of the team: coding and testing existing code in python (possibly R), data processing (cleaning, reformatting), data visualization, reading journal papers, and literature review |