Recommendation Based AI Powered Medicine Expiry and Counterfeit Detection System

Counterfeit and expired medicines pose critical global health risks, with the World Health Organization estimating that 1 in 10 medical products in developing countries are substandard or falsified. Traditional manual inspection methods are time-consuming, error-prone, and fail to scale with pharmaceutical supply chain complexity. This paper presents MediCheck, an intelligent web-based pharmaceutical authentication system that integrates Optical Character Recognition (OCR), regex-based date extraction, and deep learning to automatically detect medicine expiry dates and identify counterfeit products in real time. The expiry detection module employs Tesseract OCR with dual page segmentation modes (PSM 3 and PSM 6), combined with a layered regex engine tailored to Indian pharmaceutical labelling formats as mandated by the Central Drugs Standard Control Organisation (CDSCO), including MM/YYYY, DD/MM/YYYY, and abbreviated month-year formats. The counterfeit detection module fine-tunes an EfficientNet-B0 convolutional neural network using a two-phase transfer learning strategy, achieving a validation accuracy of 99.78% on a curated dataset of real and fake medicine images. Both modules are integrated into a Flask-based responsive web application, enabling users to upload a medicine image and receive instant verification results. The proposed system demonstrates practical applicability for pharmacies, hospitals, distribution centres, and individual patients, offering a scalable solution for reducing medication errors and protecting public health.