Peer-reviewed publications.
7 papers across healthcare AI, Arabic NLP, and LLM adaptation — including two Q1 journal articles and a top-0.1% cited Computer Science paper.
Baseline figures verified from the CV; refreshed from Google Scholar when reachable.
All papers.
- 01
Breast cancer detection using artificial intelligence techniques: A systematic literature review
Artificial Intelligence in Medicine · Elsevier (ScienceDirect)
Details428cites - 02
Arabic fake news detection based on deep contextualized embedding models
Neural Computing and Applications · Springer
Details122cites - 03
Alzheimer Detection using Different Deep Learning Methods with MRI Images
IEEE ASET — Advances in Science and Engineering Technology International Conferences · IEEE
Details12cites - 04
Death/Recovery Prediction for COVID-19 Patients using Machine Learning
International Journal of Systems Applications, Engineering & Development
Details7cites - 05
Enhancing RLHF in LLMs: Comparing BERT, XGBoost, and Deep Models for Bias Detection
IEEE DeSE · IEEE
Details0cites - 06
Text Toxicity Level Detection using Deep Contextualized Embedding Models
OkIP International Conference on Advances in Health Information Technology (AHIT) · OkIP
Details0cites - 07
Heart Failure Prediction using Machine Learning with Meta-heuristic Feature Selection
OkIP International Conference on Advances in Health Information Technology (AHIT) · OkIP
Details0cites
Currently under review.
Work submitted and awaiting peer review — listed separately from published papers.
LoRA, QLoRA, and full fine-tuning for Arabic medical question answering: a 22-condition study
Cluster Computing (Springer) · 2026
A controlled comparison of LoRA, QLoRA, and full fine-tuning across 4 open-weight models (Llama-3.1, Qwen2.5, Jais-2, Fanar) for Arabic medical QA, using a two-stage domain and task fine-tuning pipeline with multi-seed reruns, bootstrap confidence intervals, and McNemar significance testing.
Machine learning for groundwater-level forecasting (2 papers)
Under review · 2026
Two co-authored papers on machine learning for groundwater-level forecasting, covering time-series modelling and AutoML approaches.
What I research.
LLM adaptation & efficiency
Controlled studies of how large language models adapt: LoRA, QLoRA, and full fine-tuning under matched budgets, and architecture modifications that swap transformer FFN layers for learnable RBF/KAN variants across a 590M–13B scaling ladder.
Arabic & multilingual NLP
Arabic-language modelling where resources are scarce — fake news detection with contextualized embeddings, Arabic medical question answering, and bilingual Arabic/English retrieval systems.
Healthcare & clinical AI
Applied machine learning on clinical data: breast cancer detection methods, Alzheimer's detection from MRI, heart failure prediction, and COVID-19 outcome prediction.
Open to research collaboration.
Especially on LLM adaptation, Arabic and low-resource NLP, and clinical machine learning.