Chronoamperometry Data Analysis for Non-invasive Glucose Monitoring

In this project, I collaborated with a client developing a revolutionary non-invasive glucose monitoring system. Our initial focus was on calibrating the physical device to accurately convert raw signal measurements into glucose levels. We began by analyzing chronoamperometry data to identify optimal parameters for the device and establish a robust calibration method.

I took charge of developing a custom raw data parser in Python to handle the client’s proprietary data efficiently. This parser facilitated data processing and visualization, enabling insightful analysis for both myself and domain experts. Through a series of visualizations and in-depth analyses, we gained a comprehensive understanding of the device’s stability and accuracy.

Furthermore, these analyses played a crucial role in determining the best parameters for the device, such as voltage settings and sampling rates. This fine-tuning process was instrumental in ensuring the device’s reliability and precision.

Result: Our meticulous data analyses and device tuning efforts yielded high-quality data, setting a strong foundation for subsequent stages of the project. Moving forward, we planed to gather additional data and leverage machine learning techniques to further enhance the device’s performance. This involves developing ML-backed algorithms for precise calibration, ultimately enabling the transformation of raw signal readings into accurate glucose concentrations.

This project showcases my expertise in data analysis, visualization, and collaboration within the domain of healthcare technology. It demonstrates my ability to tackle complex challenges and deliver tangible solutions that drive innovation in medical device development.

Selected resources:

Artificial Intelligence | Data Science | Machine Learning | Big Data

Over two decades of comprehensive experience in data processing from BigData to AI