Our new manuscript examining the accuracy of Continuous Glucose Monitoring after meals in adults without diabetes has been published in the American Journal of Clinical Nutrition (AJCN). Summary: CGM catches postprandial trends well, however, the accuracy is lower than fingerstick meters. This discordance is important to factor in for CGM-based applications such as precision nutrition and early diabetes detection. Great Rice – Texas A&M collaboration!

Average glucose responses to meals of different macronutrient compositions using gold standard, fingerstick, and CGM measurements.

The full manuscript can be found here:

Led by our wonderful UG researcher Sahana, our abstract on data-driven insights on physical activity and hypoglycemia in type 1 diabetes has been accepted as an ORAL presentation at the prestigious American Diabetes Association (ADA) annual meeting to be held in New Orleans in June 2022. In this work, we build computation models from multi-modal data (CGM, insulin, physical activity metrics) to identify the key predictors of exercise-induced hypoglycemia. We propose a novel CGM-based measure that is independently associated with extent of hypoglycemia 2 hours after exercise. More details may be found here.

We are grateful to the Tidepool Big Data Donation project for providing the physical activity data for anlaysis. Huge thanks to the awesome Dr. Dan DeSalvo and Dr. Alex Siller for their clinical guidance, to my mentor Dr. Ashu Sabharwal for his inputs on the analysis, and to my UG mentees Sahana and Jeremiah Johnson for their fantastic work!

Multi-modal data showing glucose, physical activity, insulin, and carb intake in a study participant. Sharp glucose drop post-exercise can be observed. Image may not be reused or reshared without written permission of the authors of the abstract in the link above.

Our invited article on the equity challenge in healthcare research and how a system-wide approach is needed has been published in GetMobile journal. In this paper, we discuss four pillars to ensure successful engineering and technology solutions to health problems in underserved communities: Trust between the participants and the researchers, culturally sensitive support, locally relevant designs, and use of computational tools to harness novel insights. We therefore demonstrate that engg and tech research for these communities needs to be people-facing and guided by their challenges and requirements in order for it to be adopted and effective.
The lead author on this paper was my mentor Dr. Ashu Sabharwal and supported by myself and Dr. David Kerr, our wonderful SDRI collaborator and an expert on equitable health innovations in diabetes, The manuscript can be found here.

A participant in Sansum Diabetes Research Institute’s (SDRI) Farming for Life study wearing continuous glucose monitoring technology. Image used with permission from SDRI, all image rights reserved with SDRI.

Our latest paper titled: “The northeast glucose drift: Stratification of post-breakfast dysglycemia among predominantly Hispanic/Latino adults at-risk or with type 2 diabetes” is now published in the Lancet Eclinicalmedicine journal. In this work, we built a computational framework to analyzed the glucose response to breakfast in Hispanic/Latino adults with or at-risk for type 2 diabetes. Our results show that the breakfast glucose response has distinct patterns related to diabetes severity. This finding, once validated in larger cohorts, may potentially enable a new at-home test for diabetes progression using continuous glucose monitors (CGM).

(Distinct post-breakfast glucose patterns in individuals who are healthy but at elevated risk for type 2 diabetes, those with prediabetes, and those with type 2 diabetes.– Image copyright Souptik Barua, Dec 2021)

[Click for high-res image]

Manuscript from my PhD work on making early prediction of osteoradionecrosis in head and neck cancer patients receiving radiation treatment has been published in Frontiers in Artificial Intelligence. Our manuscript is part of a special collection on ‘Artificial Intelligence in Precision Medicine’. Paper is available to read here.

Osteoradionecrosis (ORN), or bone death, is a major side-effect of radiation treatment in a specific type of head and neck cancer patients: oropharyngeal cancer. ORN is a late complication, developing around 2 years or more after radiation treatment is complete. In this paper, we show that capturing intricate changes in CT scans in the first 6 months after treatment can predict for the eventual onset of ORN with both high accuracy and precision. We used a combination of functional PCA and random forest to quantify these fine CT scan dynamics. This work hopefully can alert physicians early on to patients who have a high risk of ORN development enabling proactive therapeutic measures MUCH BEFORE the development of ORN!

In collaboration with Sansum Diabetes Research Institute, we show that providing free vegetable prescriptions to participants with or at-risk of Type-2 diabetes improves various cardio-metabolic and quality of life factors. Using novel continuous glucose monitoring technology, we observed improvement in various glycemic metrics such as average glucose and time in 70-180 mg/dL range. A more detailed summary of the paper here:

The full paper can be found here.

Had a great time talking about my research, life as a postdoctoral research scientist, and the joys of mentoring young researchers in a Q&A with Rice’s Data to Knowledge (D2K) lab. The D2K lab provides several experiential data science courses, tailored towards building real-world data science skills. I have had a wonderful time mentoring a team of young undergraduate/1st year grad students, who won 1st prize in the fall and 2nd prize in the spring in the D2K showcase event. The full interview can be found here.

Our abstract titled A Novel Algorithm For Early Detection Of Junctional Ectopic Tachycardia In Patients With Congenital Heart Disease’ has been accepted as a poster at the prestigious 10th congress of the World Federation of Pediatric Intensive and Critical Care Societies (WFPICCS20). This work was done with a team I mentored as part of the D2K learning lab and Dr. Parag Jain at Texas Children’s Hospital. We have developed a framework for early detection of junctional ectopic tachycardia, a common form of cardiac arrhythmia in children who have had heart surgery. This framework can help clinicians respond quickly and administer life-saving treatment to patients in JET, whilst also reducing false alarms.

Thrilled to announce that I have been chosen for a Seed Fund award by PATHS-UP (https://pathsup.org/). PATHS-UP, or Precise Advanced Technologies and Health Systems for Underserved Populations , is a NSF funded engineering research center that aims to reimagine healthcare of underserved populations by developing revolutionary and cost-effective technologies and systems at the point-of-care. My award, in collaboration with the prestigious Sansum Diabetes Research Institute (SDRI), will aim to derive novel insight and prognostic biomarkers from glucose time-series data from a low-income cohort of diabetes and pre-diabetes participants.

More information here: https://pathsup.org/2020-paths-up-innovation-seed-fund-winners/