ANNU Digital Library

Photo by @inspiredimages

Communities in DSpace

Select a community to browse its collections.

Now showing 1 - 2 of 2

Recent Submissions

GENERATING SIGNALING PATHWAYS MAP FOR CARDIAC FIBROSIS FROM LITERATURE USING DEEP LEARNING
(An-Najah National University, 2026-10-09) Ayat Salman
Most diseases involve entire networks of interacting proteins rather than single proteins. Diseases typically happen because of mutations in protein sequences, or if a specific protein amount is too high or too low in a cell, or because of an issue in the interaction between the proteins in the cell if they are overacting or don’t interact at all while they should; understanding which protein-protein interactions (PPIs) are broken is therefore critical for understanding disease mechanisms and enabling targeted medicine creation. However, identifying PPI interactions using the traditional experimental methods in the labs remains computationally challenging, expensive and time-consuming. This research uses deep learning techniques that is faster and less cost to identify the PPI, it proposes a GNN architecture that uses 3D protein structures predicted using AlphaFold2 to improve PPI predictions by using pre-trained protein embeddings from D-SCRIPT. Graphs of protein interactions were built in which nodes correspond to residues, where each presented by a D-SCRIPT embedding, and edges denote proximity based on predicted intra-protein and inter-protein contacts. A learnable linear projection layer reduces dimensionality to 256, and then a 3-layer Graph Convolutional Network (GCN) encoder with residual connections and Layer normalization. A multi-head self-attention pooling mechanism generates representations at the protein level, which then feed into a 3-layer MLP classifier to make predictions. The model was trained and tested on the human PPI dataset from the STRING database, which consisted of about 150,000 protein pairs across ~15,000 unique proteins. The model implemented with a strict data leakage avoidance yielded a 96.22% accuracy, 98.81% precision, 95.37% recall, and 98.80% ROC-AUC on the held-out test set of 15,000 pairs. An ablation study was performed; comparing how well the model performed when using real AlphaFold contacts versus random synthetic contacts, leading to signify the quantification of 3D structural information. The random contacts baseline recorded an accuracy rate of 93.4%. The use of real structural contacts, therefore, enhanced the model’s accuracy rate to 96.15%, thereby validating our contention that real structural contacts do provide critical information to improve PPI prediction. On the other hand, the model performs significantly better than a purely D-SCRIPT-based model (72.33% accuracy).
IDENTIFYING THE SUCCESS FACTORS FOR AI-SUPPORTED CRISIS MANAGEMENT IN PUBLIC SECTOR IN PALESTINE: A MULTI-CRITERIA DECISION-MAKING APPROACH
(An-Najah National University, 2026-08-27) Taymaa AlSaleh
Integrating artificial intelligence (AI) into public sector crisis management poses a substantial institutional challenge, especially within resource-constrained situations with fragmented data infrastructure, underdeveloped personnel competencies, and largely absent regulatory frameworksIn order to identify and rank the key success factors (CSFs) influencing AI adoption in the Palestinian public sector, this study uses the Fuzzy Analytic Hierarchy Process (Fuzzy AHP) as its primary methodological framework, utilizing Chang's (1996) extent analysis method to calculate criteria weights. 14 domain experts from 12 Palestinian public sector organizations were given structured pairwise comparison tools; 14 of them were kept in the final analysis after the data quality was evaluated. Technical and Organizational Factors, Ethical Factors, and Training and Capacity-Building Factors are the three hierarchical dimensions in which the instrument produced 778 individual judgments among 19 sub-factors. Chang's extension analysis and geometric mean aggregation analysis resulted in a hierarchy of priorities. With 53.28% of the total important weight, Training and Capacity-Building Factors are the dominant dimension. Among the 19 sub-factors in this category, Skill Level of Personnel received the greatest priority within the Palestinian public sector (65.95% local weight; 35.14% global weight), providing clear factual proof of a key basic competency shortage. The Internet of Things infrastructure (22.72% local weight), Deep learning (21.84%), and Large-scale Machine Learning (19.80%), which together account for 64.36% of the technical category's internal weight, lead Technical and Organizational Factors, which came in second (35.15%). Reliability and Safety (18.88%) and Accountability (18.82%) were determined to be the most important governance criteria, while Ethical Factors obtained the lowest category weight (11.56%). The lack of a thorough AI regulatory framework in the Palestinian legal system is reflected in the weight of 0.00% given to regulatory training. These results, which were confirmed by sensitivity analysis in three analytical scenarios with ±20% weight perturbations, provide an evidence-based, sequentially structured framework for adoption: technological investment must be conditioned and preceded by foundational workforce development. The four pairwise comparison matrices all fulfilled the Consistency Ratio (CR) of < 0.10. The methodology is transferable to similar developing-country contexts with comparable institutional and infrastructure problems.
IMPACT OF BLEEDING-CONTROL TRAINING PROGRAM ON KNOWLEDGE AND ATTITUDES AMONG EMERGENCY NURSING STUDENTS A PRE AND POST INTERVENTION STUDY
(An-Najah National University, 2026-09-06) Mahmood Saleem Salman
THE EFFECTIVENESS OF COGNITIVE BEHAVIORAL THERAPY (CBT) PROGRAM IN REDUCING SUICIDAL IDEATION AMONG YOUTH IN PSYCHOLOGICAL DISTRESS
(An-Najah National University, 2026-08-24) Abd Elrahman Maher Jumah Abu Dalou
Choco Matic
(Dr. Mahmoud Assad, 2026-06-14) Ahmad Dardook; Sayed Qutob
The rising demand for high-quality, customized chocolate bars has driven the need for small- scale, automated production solutions. This project presents the design and implementation of a chocolate bar production line that incorporates a fully automated chocolate tempering machine and a topping system for nuts and raisins. The system integrates three microcontrollers an Arduino Mega 2560 for production line control, an ESP32 for managing the tempering machine and wireless app connectivity, and an Arduino Uno for monitoring filling levels and driving additional LCD displays. Key features include a PT100 temperature sensor for precise heating, a 1950W heating element controlled via SSR, ultrasonic sensors for topping level monitoring, and mode-selection keypads for both chocolate type and topping. Chocolate dispensing and topping systems are actuated using NEMA23 stepper motors, while a windshield-wiper motor drives the conveyor belt. Additionally, real-time feedback is provided through LCD displays, and the entire system can be operated manually or remotely via a mobile application. Thermal behavior and energy usage were modeled using fundamental heat transfer formulas, ensuring reliable tempering with realistic time estimates. The project aims to deliver a scalable, energy-efficient, and user- friendly solution suitable for artisan or semi-industrial chocolate production environments.