Longitudinal Volumetric Study for the Progression of Alzheimer's Disease from Structural MRI

TL;DR

This study uses longitudinal structural MRI data to analyze brain tissue volume changes in Alzheimer's patients, aiming to improve early detection and understanding of disease progression through advanced image processing and trend analysis.

Contribution

It introduces a comprehensive pipeline for analyzing longitudinal MRI data, combining modern preprocessing, tissue segmentation, and trend analysis to study AD progression.

Findings

Identified patterns of atrophy and tissue volume shifts in AD progression

Developed a trend analysis method using modified Mann-Kendall statistic

Provided insights into structural brain changes associated with AD advancement

Abstract

Alzheimer's Disease (AD) is an irreversible neurodegenerative disorder affecting millions of individuals today. The prognosis of the disease solely depends on treating symptoms as they arise and proper caregiving, as there are no current medical preventative treatments apart from newly developing drugs which can, at most, slow the progression. Thus, early detection of the disease at its most premature state is of paramount importance. This work aims to survey imaging biomarkers corresponding to the progression of AD and also reviews some of the existing feature extraction methods. A longitudinal study of structural MR images was performed for given temporal test subjects with AD selected randomly from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. A pipeline was implemented to study the data, including modern pre-processing techniques such as spatial image registration, skull stripping, inhomogeneity correction and tissue segmentation using an unsupervised learning approach using intensity histogram information. The temporal data across multiple visits is used to study the structural change in volumes of these tissue classes, namely, cerebrospinal fluid (CSF), grey matter (GM), and white matter (WM) as the patients progressed further into the disease. To detect changes in volume trends, we also analyse the data with a modified Mann-Kendall statistic. The segmented features thus extracted and the subsequent trend analysis provide insights such as atrophy, increase or intolerable shifting of GM, WM and CSF and should help in future research for automated analysis of Alzheimer's detection with clinical domain explainability.