Ayesha Khan
Research Subjects
Ayesha Khan has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
Alan Wang
1 shared publicationChi Sun Poon
1 shared publication- Hafiz Asad Ali
The Hong Kong Polytechnic University Hung Hom, Kowloon, Hong Kong
1 shared publication Justin Fernandez
1 shared publicationNikola Kasabov
1 shared publicationVickie Shim
1 shared publication
Latest Publications
- Open Access
- 10 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
An Interpretable Framework of Spiking neural networks for Classification and Detection of Alzheimer's Disease Using Multimodal neuroimaging data
Alzheimer's disease (AD) is a progressive brain disorder that leads to memory loss and cognitive decline, making a timely and reliable diagnosis critical. Relying on just one type of data, such as brain scans or clinical tests, often provides only a partial view of the disease. Multimodal data analysis provides a means to better understand diseases and the changes associated with brain regions by combining information from various medical imaging techniques, including structural MRI, diffusion tensor imaging, and functional MRI. The paper presents a fusion of multimodal data and its analysis using a Spiking Neural Networks (SNN) based framework, NeuCube, for detecting and classifying AD. Structural and functional data were collected using T1W, T2Flair, DTI, and fMRI images for the AD and control groups. The output confirms that NeuCube is a reliable SNN framework for modeling, classifying, and understanding multimodal data, achieving an accuracy of 88%. The results have been compared with different conventional machine learning methods to confirm the model's reliability. Our NeuCube-based model of multimodal data is efficiently implementable on neuromorphic platforms, offering both high performance and low energy use for real-time applications.
- 10 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in TechRxiv
Open AccessAn Interpretable Framework of Spiking neural networks for Classification and Detection of Alzheimer's Disease Using Multimodal neuroimaging data
Alzheimer's disease (AD) is a progressive brain disorder that leads to memory loss and cognitive decline, making a timely and reliable diagnosis critical. Relying on just one type of data, such as brain scans or clinical tests, often provides only a partial view of the disease. Multimodal data analysis provides a means to better understand diseases and the changes associated with brain regions by combining information from various medical imaging techniques, including structural MRI, diffusion tensor imaging, and functional MRI. The paper presents a fusion of multimodal data and its analysis using a Spiking Neural Networks (SNN) based framework, NeuCube, for detecting and classifying AD. Structural and functional data were collected using T1W, T2Flair, DTI, and fMRI images for the AD and control groups. The output confirms that NeuCube is a reliable SNN framework for modeling, classifying, and understanding multimodal data, achieving an accuracy of 88%. The results have been compared with different conventional machine learning methods to confirm the model's reliability. Our NeuCube-based model of multimodal data is efficiently implementable on neuromorphic platforms, offering both high performance and low energy use for real-time applications.