Guangshuai(Jerry) Han
Guangshuai Han is a fourth-year PhD candidate in Civil Engineering at Purdue University. His research focuses on utilizing AI algorithms to design materials with specialized properties, such as thermoelectric and dielectric materials. Additionally, he works on sensor design and AI-assisted signal processing for applications including structural health monitoring, wearable electronics, and pharmaceutical quality control. Guangshuai has collaborated with numerous funding sources, including the National Science Foundation, Indiana Department of Transportation, Pfizer, Eli Lilly, and Genentech.

Most of my work is currently under review, so the majority of my GitHub Pages are temporarily unavailable. They will be accessible as soon as conditions permit.

Email
Google Scholar
05 AI-Driven Moisture Content Monitoring in Pharmaceuticals

AI-Driven Moisture Content Monitoring in Pharmaceuticals
Category: Pharmaceutical Quality Control/AI 4 Science This project focuses on the development of a deep learning-guided electrochemical impedance spectroscopy (EIS) technique for monitoring the moisture content in pharmaceutical materials. The accurate determination of moisture content is crucial for maintaining the quality, stability, and efficacy of pharmaceutical products. Traditional methods often require extensive calibration, which can be time-consuming and inconsistent.

By utilizing EIS signals, I identified a strong correlation between the electrical properties of pharmaceutical materials and their moisture content. An equivalent circuit model was employed to understand the underlying mechanisms, providing valuable insights into the sensitivity of EIS to moisture variations.

To further enhance the accuracy and efficiency of moisture content monitoring, I incorporated a 1D convolutional neural network (1DCNN) model into the EIS data processing pipeline. This model demonstrated a predictive accuracy with an average error as low as 0.69%, making it a pioneering study in the application of deep learning for real-time, calibration-free moisture content monitoring in pharmaceuticals.

The developed method not only offers a reliable solution for pharmaceutical quality control but also has potential applications in other industries, including food, energy, environmental monitoring, and healthcare. This work represents a significant step forward in the integration of AI with traditional sensing techniques to improve manufacturing processes.


Github Page:
EIS Processing: EIS_Moisture




INDEX  NEXT
©Guangshuai Han 2024
All information and images on this website are the exclusive property of the owner and may not be used for any other purpose without permission.