Rapid detection and quantification of falsified Viagra using cloud-based portable NIR technology and machine learning

Area of Science:

• Analytical Chemistry
• Pharmaceutical Sciences
• Spectroscopy

Background:

• Falsified medications pose a significant global health risk, driven by increased globalization and online sales.
• The illicit trade in counterfeit drugs is highly profitable with minimal risk, exacerbating the problem.

Purpose of the Study:

• To develop and validate a portable Near-Infrared (NIR) spectroscopy method for rapid, on-site detection and quantification of falsified Viagra tablets.
• To integrate this technology with a cloud-based data processing architecture for enhanced accessibility and efficiency.

Main Methods:

• Utilized portable NIR spectroscopy for non-destructive analysis of authentic and falsified Viagra tablets.
• Employed Principal Component Analysis (PCA), Euclidean distance, and Support Vector Machine (SVM) for sample classification.
• Developed machine learning models for accurate quantification of sildenafil content, validated against a chromatographic reference method.

Main Results:

• Portable NIR spectroscopy effectively differentiated authentic from falsified Viagra tablets.
• The developed models achieved high accuracy in classifying samples and quantifying sildenafil content.
• The system demonstrated potential for seamless integration into online platforms and mobile applications for field use.

Conclusions:

• Portable NIR spectroscopy, coupled with advanced data processing, provides a rapid, reliable, and non-destructive solution for combating falsified medications.
• This technology empowers regulatory bodies and healthcare providers to ensure drug quality and safety in real-time.
• The findings highlight a transformative tool for field deployment in pharmaceutical quality control and public health protection.