Spring 2020 - Development of Statistical Methods and Visualization Tools for Analysis of Continuous Glucose Monitoring (CGM) Data

Affiliations:	DeBakey Executive Research Leadership Program
Project Leader:	Dr. Irina Gaynanova irinag@stat.tamu.edu Statistics
Faculty Mentor:
Meeting Times:	To be determined by team
Team Size:	6 (Team Full)
Open Spots:	0
Special Opportunities:	earn 491 credit flexible schedule besides the required meeting time potential of co-authorship on developed software products and resulting publications
Team Needs:	Attend the confirmed weekly meeting time; there will be other meeting times by need base. We are looking for students with a diverse range of skills who want to be part of a team to accomplish this research (different skills are expected from different team members, overlap is possible). Team members are needed for the following project components: searching the web and scraping the found datasets from the web, putting them in a ready format for analysis (we are looking for individual(s) with experience in web scraping and familiarity with different data formats, familiarity with R language is preferred but not required) statistical analysis of the data (we are looking for individual(s) with experience in statistical analysis and data processing/filtering using R language) visualization of the data (we are looking for individuals(s) with experience in data visualization, specifically data over time) identifying gaps and opportunities to contribute to diabetes CGM literature and create a comprehensive literature review (we are looking for individual(s) who are familiar with diabetes-related terminology and are excited about reading diabetes-related literature pertaining to CGM).
Description:	Continuous glucose monitors (CGM) are small wearable devices that allow to measure the glucose levels continuously throughout the day, with some monitors taking measurements as often as every 5 minutes. In addition to providing patients with diabetes with frequent alerts on the status of their glucose levels, CGMs can supply researchers and clinicians with a wealth of data that has the potential to improve the diabetes management. However, while CGMs measure the time-dynamic of glucose profile and play an increasing role in clinical practice, their measurements are highly dependent on environmental and behavioral factors (i.e., meals, physical activity) that are often unknown. As a result, these data are often analyzed using very crude statistical measures (such as mean) that fail to fully explain clinical measures such as Hemoglobin A1C. The purpose of this project is to assemble a large collection of publicly available CGM data for the purpose of comparing different statistical metrics and their associations with clinical outcomes, as well as developing new metrics based on the identified needs from the literature. The project is very flexible and the specific directions will be driven by the team developments. This is an early-stage project that has potential to expand in scope over several semesters based on interests of participants.