Volatile Organic Compound Profiler

For Use in Future Glucose Monitoring

Presenter: Anthony Arjona Pech

Co-Presenter(s):
Jonathan Cervantes, Jessica Mellor

Presenter Status: Undergraduate student

Academic Year: 20-21

Semester: Spring

Faculty Mentor: Sudhir Shrestha

Department: Engineering

Funding Source/Sponsor: Other

Other Funding Source/Program: Student Research Award

Screenshot URL: https://drive.google.com/uc?id=1dTJbhSKotG3jJuhoIOL_7U67EnYKaV1q

Abstract:
Volatile organic compounds (VOCs) have been shown to appear in the exhaled breath of an individual albeit in minuscule quantities. What researchers have found, however, is that the concentrations of specific VOCs found in our breath can reveal useful information regarding a person’s physiological state. A potential use for this could be the future research and development of non-invasive glucose monitoring systems. The issue, however, is the difficulty to accurately measure the concentrations of VOCs found in a breath sample. To aid with the measurement and profiling of VOCs, machine learning algorithms are used with Tedlar bags. This has the disadvantage of being expensive, time-consuming, and overall inefficient as it requires a very specific set of expensive tools in a lab setting. There needs to be a way to train algorithms for use in VOC profiling in a way that is relatively inexpensive, easy to use, and efficient in that it is not dependent on a patient being in a typical lab setting. Our proposed solution is a small handheld device capable of accepting a trained machine learning algorithm, displaying results on a user-friendly graphical user interface (GUI), and saving data on an external memory card, all enclosed in a 3D printed case. More specifically our device includes a demonstration algorithm that is trained to be able to differentiate the VOC profiles found in Equate and Germ-X brands of hand sanitizers in order to correctly identify the brand used. The purpose of this demonstration algorithm is to set up the basic framework of algorithm implementation to help to show the mobility and ease of use of our system and will contribute to the development of sophisticated machine learning algorithms for use in VOC profiling, which in itself will allow for further developments of non-invasive medical solutions for glucose monitoring. To accomplish our goal we are using an STM32 development board, an LCD display, three Figaro VOC sensors, and a machine learning algorithm that we are developing based on Keras. Currently, we have successfully implemented all subsystems and configured them to work with one another to form our device. Specifically, we have created a fully functional multi-page GUI that can be navigated via a D-pad and communicate with other peripherals, implemented an SD card module, added an external battery power source that makes use of a boost converter to power all peripherals, created a standardized testing method using our VOC sensors, and integrated our machine learning algorithm, Profilemaster. We’ve also designed a 3D printed case, and have developed several custom printed circuit boards (PCBs) to house all these components. The GUI is capable of formatting and saving sensor data onto an SD card, initiating and capturing sensor data, and displaying past readings, a timer and buzzer system have also been implemented to help with creating a standardized testing procedure. We have enhanced our sensor data gathering technique by implementing a standardized data collection procedure, and have used that data to help train our algorithm. Our current algorithm is trained to operate with an accuracy of 76%, more data needs could be processed for us to reach a higher level of accuracy but we have met our target accuracy of 75%. We have successfully imported this iteration of our algorithm into the ST IDE and configured our program to take real-time sensor data as the input, and output a classification in real-time through the GUI. Ultimately, we have accomplished our goal for this project and have put ourselves in a position to easily make adjustments and improvements as needed.