fairfolio
Saved Projects
← Back to Explore

GlucoseAssist: A Novel, Personalized System for Prediction of Blood Glucose Levels and Early Identification of Dysglycemic Events using Artificial Intelligence

JSHS · 2023

Prisha Shroff, Mrs. Debbie Nipar, Dr. Hassan Ghasemzadeh · Hamilton High School, Chandler, AZ

Overview

Blood glucose (BG) control is important for all individuals, especially diabetics, to evaluate and manage their metabolic health. Poor BG control results in dysglycemia. Frequent exposure to dysglycemia leads to cardiovascular disease, seizures, loss of consciousness, and potentially death. Today, many individuals struggle with BG control due to a multitude of interrelated behavioral, physiological, and biological factors such as food, insulin intake, and metabolism rate. There is a need for a solution that can accurately predict future BG levels, and dysglycemic events. However, current research uses limited input parameters, lacks potential meal-based predictions and is data-hungry and computationally expensive. In this research, GlucoseAssist, a novel, personalized, AI-driven system, was developed to predict BG response in real-time and identify dysglycemic events based on diet, health, and medication data. Importantly, the devised system identifies the timing of the impending health events and provides preventative measures. The architecture uses a multimodal convolutional neural network and random forest classifier with time series data from a clinical dataset with 20,040 Continuous Glucose Monitor records. GlucoseAssist accurately predicts the BG response for the next 30 minutes with a Root Mean Squared Error of 1.230, nominal Mean Absolute Error of 0.920, and accuracy of 97.07% for identification of dysglycemic events. GlucoseAssist demonstrates the feasibility of developing a data-driven solution that makes accurate, personalized predictions with a limited training dataset size. GlucoseAssist has the potential to positively impact over 911 million individuals’ lives, who need a personalized, cost-efficient solution to help with BG control.

Competition history

  • JSHS 2023 Category not listed

Resources

Related projects

Closest projects by meaning, across every fair and year in the corpus.

Source: Junior Science and Humanities Symposium

Save projects to your library

Sign in with Google to keep track of projects you find interesting, organized into folders. An account also raises your daily allowance for “Has this been done?”, and lets you create a key for the MCP server with a much higher limit than anonymous use. Browsing stays public.

Continue with Google