Journal of Industrial Engineering and Management Studies

Abstract In the landscape of healthcare, ensuring the accuracy and security of prescription processes is crucial for maintaining patient safety and upholding ethical standards. This paper presents a novel biometric authentication framework designed to address the vulnerabilities in traditional authentication methods such as passwords and codes, which are prone to misuse. By integrating fingerprint and iris recognition, the proposed multi-modal system provides a robust solution to prevent unauthorized access to prescription data. This study collected biometric data from 600 doctors, comprising 600 fingerprint images and 1200 iris images, to rigorously evaluate the system’s performance. Detailed information about the CNN architecture, including layers, activation functions, and loss functions, is provided. The model's effectiveness was measured using comprehensive metrics such as accuracy, precision, recall, and F1-Score, demonstrating a significant improvement over existing methods. Furthermore, a statistical analysis was conducted to verify the reliability of the results, with comparisons drawn against baseline methods. The findings underscore the importance of enhancing biometric authentication systems and contribute to the development of secure and reliable identity verification solutions across the healthcare sector. This research not only bolsters the security of prescription processes but also reinforces the ethical principles guiding medical practice, offering a significant step forward in preventing fraud in healthcare systems.