Sparse kernel entropy component analysis for dimensionality reduction of neuroimaging data.

Area of Science:

• Neuroscience
• Machine Learning
• Biomedical Imaging

Background:

• Neuroimaging data is high-dimensional, necessitating dimensionality reduction for feature extraction.
• Kernel Entropy Component Analysis (KECA) is effective but computationally inefficient for large datasets.

Purpose of the Study:

• To develop an efficient dimensionality reduction algorithm for neuroimaging data.
• To improve the classification accuracy of Alzheimer's disease (AD).

Main Methods:

• Proposed a Sparse Kernel Entropy Component Analysis (SKECA) algorithm using a recursive divide-and-conquer approach.
• Applied SKECA for dimensionality reduction of neuroimaging data for AD classification.
• Compared SKECA against KECA, PCA, KPCA, and sparse KPCA.

Main Results:

• SKECA demonstrated superior performance in extracting discriminative features compared to other methods.
• The proposed SKECA algorithm significantly improved Alzheimer's disease classification accuracy.

Conclusions:

• SKECA offers a computationally efficient and effective solution for dimensionality reduction in neuroimaging.
• SKECA shows great potential for improving the diagnosis of Alzheimer's disease using neuroimaging data.