Non-Invasive Detection of Expired Medicines Using Artificial Intelligence and Image Analysis

Department of Computer Science & Engineering, Institute of Engineering and Management (IEM), University of Engineering and Management Kolkata, IEM Centre of Excellence for InnovAI, Kolkata, West Bengal, India

^* Corresponding Author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

This study investigates non-invasive techniques for determining whether drugs are safe, expired, or deteriorating using packaging images. This method uses the extraction of visual and text characteristics driven by artificial intelligence (AI) as opposed to conventional methods for manipulation, chemical analysis or barcode scanning. Optical Character Recognition (OCR) is used to identify expiry-related text, to designate image features using the vector neural network (CNN) and to designate relationships to the graph neural network (GNN). In addition to self-clicking test photos, an image data set for pharmaceutical tablet packs issued. The model obtained accuracy (0.91), accuracy (0.91), recall (0.89) and F1 score (0.90). The results show that the packaging characteristics have sufficient patterns to accurately assess the quality of drugs and allow safer and more scalable quality control without destroying seals.