Publications · Methods · Evidence
I am Muhammad Hamza, an AI Engineer and research co-author working across medical imaging and machine learning. Each paper has a 12-part case study, a methodology flowchart, reported results and a clear account of my implementation and experimental work.
Updated · Google Scholar · ORCID
Vehicular Communications · 2026
A hybrid intrusion detector brings sparse network attributes and temporal traffic context together through cross-attention. This publication studies adversarial robustness and minority-attack detection.
Ghulam Ali, Mohd Khalid Awang, Javed Rashid, Muhammad Hamza, Khalid Yahya, Anwar Ghani
Ain Shams Engineering Journal · 2025
3D-MobiBrainNet examines multi-class Alzheimer's classification from volumetric brain MRI. Its multi-plane design preserves complementary anatomical views for distinguishing Alzheimer's disease, mild cognitive impairment and cognitively normal cases.
Zia-ur Rehman, Mohd Khalid Awang, Ghulam Ali, Muhammad Hamza, Tariq Ali, Muhammad Ayaz, Mohammad Hijji
IEEE Access · 2026
WIDENet studies efficient segmentation of skin lesions in dermoscopic images. The focus is the lesion boundary, including fine structures that can disappear during conventional downsampling.
Ghulam Ali, Mohd Khalid Awang, Javed Rashid, Muhammad Hamza, Osama A. Khashan, Anwar Ghani
Ain Shams Engineering Journal · 2026
This paper investigates five-class diabetic retinopathy grading from retinal images. It addresses training instability, imbalanced classes and visually similar disease stages.
Ghulam Ali, Mohd Khalid Awang, Tariq Ali, Mohammad Hijji, Muhammad Ayaz, Muhammad Hamza, Azqa Jafar