fMRI-based Alzheimer's disease detection via functional connectivity analysis: a systematic review

Area of Science:

• Neuroscience
• Medical Imaging
• Artificial Intelligence

Background:

• Alzheimer's disease (AD) is a leading cause of dementia and memory loss globally.
• Early diagnosis of AD is critical for timely patient care and symptom management.
• Non-invasive imaging techniques like fMRI are vital for exploring brain connectivity.

Purpose of the Study:

• To systematically review Alzheimer's disease detection methods using functional connectivity from fMRI datasets.
• To analyze the machine learning pipeline for AD prediction, from data preprocessing to algorithm application.
• To identify current trends and future research directions in medical imaging for AD detection.

Main Methods:

• Systematic literature review of fMRI studies on Alzheimer's disease detection (2018-2024).
• Analysis of functional connectivity features derived from fMRI data.
• Evaluation of machine learning and deep learning algorithms applied to AD prediction.
• Examination of data preprocessing, feature computation, extraction, and selection techniques.

Main Results:

• Functional connectivity analysis of fMRI data shows promise for Alzheimer's disease detection.
• Machine learning and deep learning models are increasingly utilized for predicting AD occurrence.
• The review synthesizes findings on the efficacy of various analytical approaches.

Conclusions:

• Accurate and efficient early detection of Alzheimer's disease remains a significant challenge.
• Advancements in machine learning and medical imaging offer potential for improved AD diagnosis.
• Further research is needed to refine detection methods and overcome clinical hurdles.