International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 06 | Jun 2025
p-ISSN: 2395-0072
www.irjet.net
Deep Learning for Medical Lab Procurement - CNN, VGG16, and MobileNet in Secure Face Authentication and Inventory Tracking E.Sumalatha
Manoj Yadav
Md. Aftab Alam
Computer Science Computer Science and and Engineering Engineering Shri Venkateshwara Shri Venkateshwara University University Gajraula, Uttar Gajraula, Uttar Pradesh 244236 Pradesh 244236 ------------------------------------------------------------------------***------------------------------------------------------------------------M.Tech, Computer Science and Engineering Shri Venkateshwara University Gajraula, Uttar Pradesh 244236
Abstract— Considering the reality that healthcare
medical equipment companies, reducing inefficiencies, delays, and manual errors [1]. Traditional processes often suffer from poor communication and lack of accountability, which this automation aims to resolve through a centralized, transparent platform [2]. By introducing automated workflows, email notifications, and secure face authentication, the system enhances efficiency, security, and collaboration among stakeholders [3]. The goal is to ensure seamless communication, accurate tracking, and scalable operations, enabling organizations to meet current and future demands effectively [4].
supplies are essential to deliver adequate health care and experiences considerable costs for the medical field, very has been learned about what is needed to get it economically. This work focuses on streamlining and automating the procurement and tracking process of laboratory items. This automation improves efficiency, reduces manual errors, and ensures seamless communication between departments via email notifications. Our system incorporates a face authentication feature for enhanced security. During user registration, users can register their faces, and when they attempt to log in, the system checks the live image against the registered face. Only if the live face matches the registered one, the user is granted access; otherwise, they are denied login. The system uses Convolutional Neural Networks (CNNs), VGG16, and MobileNet for live face recognition, ensuring secure access by comparing the live image with the registered face. These models help verify the user's identity and improve security, as they rely solely on live face detection and not other images. Finally, we validate the performance in means of several performance measures-viz., precision, recall, F1score, and support for the three models CNNs, VGG16, and MobileNet across certain classes like class 0, class 1, accuracy, macro avg., and weighted avg. Best precision value of 81%; best recall value of 81%; and best F1-score value of 79% were obtained for the two methods, namely, CNN and MobileNet, whereas VGG-16 method produced the least precision value of 76%; least recall value of 75%; and least F1-score value of 77%.
In our paper, we aim to revolutionize the traditional procurement and tracking processes by introducing a fully automated, secure, and efficient system. In the medical equipment industry, where timely procurement and accurate tracking of laboratory items are critical, manual processes often lead to delays, errors, and communication gaps among stakeholders such as indenters, purchase departments, suppliers, stores, and administrators. This paper leverages a web-based platform to automate workflows, enabling users to raise purchase requests, monitor approvals, and track inventory seamlessly. Key features include automated email notifications, centralized ledger updates, and streamlined documentation with Goods Receipt Notes (GRN) and Material Issue Vouchers (MIV). To ensure robust security, the system integrates advanced face authentication using Convolutional Neural Networks (CNNs), VGG16, and Mobile Net, restricting access to authorized personnel only. By enhancing transparency, accountability, and operational efficiency, this paper addresses critical challenges in the procurement lifecycle while laying the foundation for scalable and future-ready solutions. It ultimately contributes to improved organizational productivity and ensures uninterrupted operations in the medical equipment sector.
Keywords- Lab item tracking, Medical equipment, Procurement process, Goods Receipt Note (GRN), and Material Issue Voucher (MIV).
I. INTRODUCTION The motivation for this paper stems from the need to streamline the procurement and tracking of lab items in
© 2025, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 928