11/08/2023
By Nazim Ahmed Belabbaci

Master’s thesis defense in Computer Science

Candidate Name: Nazim A. Belabbaci
Date: Wednesday, Nov. 22, 2023
Date: 1 p.m.
Location: TBD and via Zoom

Thesis Title: A Prototype Smartwatch Utilizing Spectroscopy for Real-Time Monitoring of Hydration Status

Committee Members: 

  • Advisor Mohammad Arif Ul Alam, Assistant Professor, Miner School of Computer & Information Sciences
  • Jerome Delhommelle, Associate Professor, Chemistry Department, Kennedy College of Science
  • Hong Yu, Professor, Miner School of Computer & Information Sciences

Abstract:

The demand for continuous health monitoring solutions has led to the development of innovative wearable biosensors. In this thesis, we introduce a novel approach to real-time hydration assessment using smartwatches equipped with a low-cost spectroscopy sensor. By integrating this technology into an everyday wearable, we aim to provide a convenient and non-invasive method for monitoring hydration levels based on blood electrolytes concentration.

We present two significant use cases: 1.The measurement of electrolyte solutions using our low-cost spectroscopy sensor and benchmark it with a high-resolution spectrometer that follows industry standards. 2. The assessment of skin hydration during a workout and fasting experiments. These use cases demonstrate the credibility of the proposed system.

We describe the signal processing techniques we used to extract meaningful data from spectroscopic measurements. Additionally, an AI algorithm is implemented on the edge allowing real-time classification of hydration status into three distinct classes.

In the results evaluation section, we present the findings of our research, showcasing the system's accuracy and performance in assessing hydration status. We also delve into additional results, focusing on emotion recognition. For this purpose, a dedicated experimental setup is described, involving the use of spectroscopy data to develop an algorithm to classify emotions as sad or happy.

In conclusion, our thesis underscores the significance of smartwatch-based electrolyte measurement for real-time hydration assessment and its potential applications in diverse areas of health monitoring. We discuss the implications of our findings and suggest future work that can further enhance this technology's capabilities.