Honors Theses

Date of Award

12-2024

Document Type

Undergraduate Thesis

Degree Name

BS

Department

Electrical and Computer Engineering

Faculty Mentor

Na Gong

Advisor(s)

Mohamed Shaban, Samuel Russ

Abstract

Diabetes Mellitus (DM) refers to a set of metabolic disorders which are generally characterized by an inability to produce or use insulin, a hormone which enables the metabolization of glucose in the blood. Blood glucose (BG) levels are an important health indicator in the management of DM. Conventional methods of measuring BG use direct blood samples and electrochemical analysis. While accurate, these ‘invasive’ methods can cause discomfort, tissue damage, and pose a risk of infection for the patient. Developing a truly ‘noninvasive’ method of measuring or approximating BG has grown increasingly relevant as technology advances and cases of DM continue to rise. In particular, recently, AI-empowered techniques have shown promising results in BG prediction. However, existing techniques are costly and usually need support from the internet, smart phones, or a backend server/cloud. This thesis designs, develops, and prototypes an AI-empowered wearable device (WD) which combines two methodologies for estimating BG and explores the validity of having such estimations performed on a purely edge-based device. The BG estimation methods explored for this device are photoplethysmography (PPG) and breath acetone measurement, with an AI model used to make estimations based on this data.

Comments

© 2024 Luke Young ALL RIGHTS RESERVED

Share

COinS