Published Paper Details:

The widespread circulation of counterfeit pharma-ceutical products poses a critical threat to public health, patient safety, and the integrity of medicine supply chains worldwide. Sophisticated counterfeit products are visually indistinguishable from genuine medicines, as illustrated in Fig. 1, making manual verification by consumers and pharmacists practically infeasible at scale. Existing verification methods that rely on manual packaging inspection or trust-based purchasing are inherently unreliable and inadequate to detect modern counterfeiting techniques. This paper presents a comprehensive Fake Medicine Detection System leveraging QR code technology, integrating role-based access control, real-time UUID-based verification, and intelligent administrative analytics within a unified web platform. The proposed architecture comprises three functional modules: a Mobile Application Frontend enabling QR code scanning and verification; a Backend Processing System executing QR validation, authentication logic, and the Fake Detection Algorithm; and an Admin Dashboard providing monitoring, reporting, and counterfeit trend analysis. A centralized Identity and Access Management (IAM) layer enforces Role-Based Access Control (RBAC) across consumer and administrator roles. Each medicine batch is assigned a Universally Unique Identifier (UUID) upon registration, encoded into a cryptographically secured QR code, and stored in a centralized database. Upon scanning, the system performs real-time UUID validation through a secure Verification API and returns an immediate authenticity verdict-- Genuine or Fake--with full batch metadata. Training and validation accuracy curves confirm stable model convergence, and confusion matrix analysis validates balanced classification without systematic class bias. Evaluation on 500 simulated scan events demonstrates 99.2% overall verification accuracy with 106 ms average end-to-end latency, demonstrating the efficacy of the proposed platform for pharmaceutical safety.