VOC-Based ML Cancer Diagnosis Machine
School Name
Academic Magnet High School
Grade Level
10th Grade
Presentation Topic
Biochemistry
Presentation Type
Non-Mentored
Abstract
This project aims to develop a portable and cost-effective solution for volatile organic compound (VOC) detection, focusing on the detection of 2-butanone, a potential biomarker for certain medical conditions. Leveraging the Raspberry Pi Zero W platform and VOC sensors the system integrates signal processing techniques and machine learning algorithms for accurate and real-time analysis of VOC concentrations. The implementation includes the utilization of an Analog-to-Digital Converter (ADC) to interface sensors with the Raspberry Pi for analog signal acquisition. The collected sensor data undergoes preprocessing to extract relevant features, followed by the application of a Random Forest machine learning model, trained on existing datasets sourced from the internet. The system employs Python for scripting and data analysis, enabling seamless integration with the Raspberry Pi environment. Through the utilization of portable hardware and robust software methodologies, this project seeks to provide a versatile solution for VOC detection, with potential applications in medical diagnostics, environmental monitoring, and indoor air quality assessment. The scalability and adaptability of the proposed system offer opportunities for customization and further development, catering to diverse use cases and addressing emerging challenges in VOC detection and analysis. Overall, this project contributes to advancing the field of VOC sensing technology, offering a promising avenue for enhanced healthcare and environmental monitoring solutions.
Recommended Citation
Hamilton, Francesco, "VOC-Based ML Cancer Diagnosis Machine" (2024). South Carolina Junior Academy of Science. 531.
https://scholarexchange.furman.edu/scjas/2024/all/531
Location
RITA 365
Start Date
3-23-2024 11:30 AM
Presentation Format
Oral and Written
Group Project
No
VOC-Based ML Cancer Diagnosis Machine
RITA 365
This project aims to develop a portable and cost-effective solution for volatile organic compound (VOC) detection, focusing on the detection of 2-butanone, a potential biomarker for certain medical conditions. Leveraging the Raspberry Pi Zero W platform and VOC sensors the system integrates signal processing techniques and machine learning algorithms for accurate and real-time analysis of VOC concentrations. The implementation includes the utilization of an Analog-to-Digital Converter (ADC) to interface sensors with the Raspberry Pi for analog signal acquisition. The collected sensor data undergoes preprocessing to extract relevant features, followed by the application of a Random Forest machine learning model, trained on existing datasets sourced from the internet. The system employs Python for scripting and data analysis, enabling seamless integration with the Raspberry Pi environment. Through the utilization of portable hardware and robust software methodologies, this project seeks to provide a versatile solution for VOC detection, with potential applications in medical diagnostics, environmental monitoring, and indoor air quality assessment. The scalability and adaptability of the proposed system offer opportunities for customization and further development, catering to diverse use cases and addressing emerging challenges in VOC detection and analysis. Overall, this project contributes to advancing the field of VOC sensing technology, offering a promising avenue for enhanced healthcare and environmental monitoring solutions.
