An efficient hybrid approach for diagnosis high dimensional data for Alzheimer's diseases using machine learning algorithms

Abstract EN

Alzheimer’s disease (AD) is the most familiar type of dementia, a well-known term for memory loss and other cognitive disabilities.

the disease is dangerous enough to interfere with ordinary life.

identifying AD in the early stages remains an extremely challenging task, meanwhile, the progression of it develops several years before observing any symptoms.

the fundamental issue addressed during diagnosis is the high dimensionality of data.

however, not all features are relevant for solving the problem, and sometimes, including some irrelevant ones may deteriorate the learning performance.

therefore, it is essential to do feature reduction by selecting the most relevant features.

in this work, a hybrid approach random forest partial swarm optimization (RF-PSO) for high-dimensional feature selection is proposed.

the fundamental reason behind this work is to support geriatricians diagnose AD ; by creating a clinically translatable machine learning approach.

the dataset created by Alzheimer's disease neuroimaging initiative (ADNI) was used for this purpose.

the ADNI dataset contains 900 patients whose diagnostic follow-up is available for at least three years after the baseline assessment.

the reason behind choosing is their strength in solving large-scale optimization problems with high data dimensionality.

the experiments show that RF-PSO outperforms most of the others found in the literature.

it achieved high performance compared to them.

the accuracy rate of this approach reached 95% for all the AD stages.

in a comparison with random forest which achieve 86%, while partial swarm optimization got 89%.