Identifying regions of interest for discriminating Alzheimer's disease from mild cognitive impairment

Abstract

Alzheimer's disease (AD) is one of the most common types of dementia that affects elderly people, with no known cure. Early diagnosis of this disease is very important to improve patients' life quality and slow down the disease progression. Over the years, researchers have been proposing several techniques to analyze brain images, like FDG-PET, to automatically find changes in the brain activity. This paper compares regions of voxels identified by an expert with regions of voxels found automatically, in terms of corresponding classification accuracies based on three well-known classifiers. The automatic identification of regions is made by segmenting FDG-PET images, and extracting features that represent each of those regions. Experimental results show that the regions found automatically are very discriminative, outperforming results with expert's defined regions.

Classification of Alzheimer’s and MCI Patients from Semantically Parcelled PET Images: A Comparison between AV45 and FDG-PET

“…The results obtained in this study using the ADNI-PET images were similar to previous works [ 9 , 20 ], which also used VI for feature-learning. This confirms the high diagnostic power of uptake values in discriminating between different cognitive stages, particularly by integrating a progression component using time series features to the analysis.…”

“…In a work by Rodrigues and Silveira using the same longitudinal database, the best classification accuracy was obtained by combining two timepoints (baseline and 12-month follow-up) (93% for NC versus AD) [ 19 ]. In another study, Aidos et al used VI in some ROIs which were manually segmented by experts [ 20 ]. By using an automated feature selection method to classify the subjects, they used mutual information to rank the features and identify the highest values to reduce the number of features.…”

“…The main objective is to develop a method to select the optimal features using PET images alone to classify AD, EMCI, LMCI, and NC. The framework is evaluated with a large sample size of 660 subjects, in contrast to previous works which used a more limited number of subjects (under 250) [ 9 , 19 , 20 ]. Furthermore, while previous methods were proposed to predict the conversion [ 22 , 23 ], they were designed for whole-brain processing and did not focus on particular regions prone to amyloid accumulation.…”

Predictive Biomarker Grade Transfer Alzheimer's disease and Mild Cognitive Impairment

“…In 2014, Alzheimer's disease affecting mainly aged persons, many researchers have been implementing different techniques to analyze brain images like FDG -PET, as a result, shows automatically very discriminative results [3]. In 2015, Speech Prosody-Based Cognitive Impairment Rating assessment (SPCIR) can discern the difference between the MCI and AD utilizing prosodic sign during the investigation test [8].…”

“…MCI is the phase among normal aging predictable natural aging cognitive decline and the more extreme decline in dementia. It can include memory, vocabulary, thinking, decision issues, and language that are greater than the changes associated with a normal aged person [1][2][3]. The person with MCI may be aware of having 'slipped' memory or mental functions.…”