On Monday, August 4, 2025, master's student Zainab Hussein Laibi was defended and awarded a master's degree for her thesis entitled

"Design and Implementation of a Fully Computerized Eye-Care Management System"

Under the supervision of lecturer Dr. Atheel Nawfal Mohammed Taher.

The defense committee consisted of:

Prof. Dr. Mahmoudski Abdullah, Chairman

Assistant Professor Dr. Israa Badr Nasser, Member

Dr. Ruslan Saad Abdulrahman, Member

This thesis presents the design, development, and evaluation of a Fully Computerized Eye-Care Hospital Management System (ECHMS), developed to improve the efficiency, accuracy, and quality of services in eye-care institutions, with implementation to Ibn Al-Haytham Eye Hospital. The system integrates a modular web-based hospital management interface with an advanced diagnostic image processing module aimed at automating administrative operations and enhancing early detection of Diabetic Macular Edema (DME).

The methodology comprises two primary components: (1) a web-based management system developed using HTML, CSS, JavaScript, PHP, and MySQL, which handles user authentication, patient registration, appointment scheduling, electronic health records, and image uploading; and (2) an automated diagnostic module developed in Python using deep learning and transfer learning techniques for image classification. The image processing pipeline includes preprocessing (noise reduction, contrast enhancement, normalization), feature extraction using a pre-trained Convolutional Neural Network (CNN) model via transfer learning, and classification using a rule-based algorithm designed with clinical criteria to distinguish between normal and DME-affected retinal images.

A dataset of over 800 high-resolution retinal fundus images—labeled by expert ophthalmologists—was used to train and validate the diagnostic model. The system was evaluated on a subset of 160 images, yielding an overall classification accuracy of 97%, with high sensitivity and specificity. Additionally, system usability was assessed by five domain experts, who rated the user experience and functionality with an average score of 9.5 out of 10.

The proposed system effectively streamlines clinical workflows, reduces manual errors, and supports timely and accurate diagnosis of DME. Its scalable architecture and integration of AI-driven diagnostics represent a significant step toward digital transformation in ophthalmology. Future enhancements include multi-disease detection, mobile platform integration, and incorporation of real-time imaging and telemedicine capabilities.