Salleh Sonko

Research Interests

My research interests lie in the application of deep learning and machine learning techniques to biomedical signal analysis, with a particular focus on EEG-based diagnosis of neurodegenerative disorders such as Alzheimer’s Disease (AD) and Frontotemporal Dementia (FTD). I investigate how different signal representations—including time–frequency transforms, functional connectivity measures, and minimally processed raw EEG—affect the ability of compact convolutional neural networks to learn clinically meaningful patterns. A central theme of my work is designing efficient and interpretable models that can reliably distinguish between neurological conditions while maintaining methodological rigor.

My research also emphasizes the importance of robust evaluation protocols, particularly strict subject-wise validation, to ensure realistic estimates of model generalization in clinical settings. In addition, I incorporate explainable AI techniques such as Grad-CAM to better understand how deep learning models identify physiologically relevant EEG features associated with cognitive decline. More broadly, my interests include biomedical signal processing, representation learning, interpretable artificial intelligence, and the development of reliable machine learning frameworks for healthcare applications.

Publications

1. Scoping the Landscape of Deep Learning for Alzheimer’s Disease Stage Classification: Methods, Challenges, and Opportunities

2. Analyzing Insider Cyber Threats and Human Factors within the Framework of Agriculture 5.0

3. Society 5.0 enabled agriculture: Drivers, enabling technologies, architectures, opportunities, and challenges

4. Predicting Long-Term Type 2 Diabetes with Artificial Intelligence (AI): A Scoping Review