Scoring by nonlocal image patch estimator for early detection of Alzheimer's disease

Abstract

Detection of Alzheimer's disease (AD) at the first stages of the pathology is an important task to accelerate the development of new therapies and improve treatment. Compared to AD detection, the prediction of AD using structural MRI at the mild cognitive impairment (MCI) or pre-MCI stage is more complex because the associated anatomical changes are more subtle. In this study, we analyzed the capability of a recently proposed method, SNIPE (Scoring by Nonlocal Image Patch Estimator), to predict AD by analyzing entorhinal cortex (EC) and hippocampus (HC) scoring over the entire ADNI database (834 scans). Detection (AD vs. CN) and prediction (progressive — pMCI vs. stable — sMCI) efficiency of SNIPE were studied using volumetric and grading biomarkers. First, our results indicate that grading-based biomarkers are more relevant for prediction than volume-based biomarkers. Second, we show that HC-based biomarkers are more important than EC-based biomarkers for prediction. Third, we demonstrate that the results obtained by SNIPE are similar to or better than results obtained in an independent study using HC volume, cortical thickness, and tensor-based morphometry, individually and in combination. Fourth, a comparison of new patch-based methods shows that the nonlocal redundancy strategy involved in SNIPE obtained similar results to a new local sparse-based approach. Finally, we present the first results of patch-based morphometry to illustrate the progression of the pathology.

Anatomically Constrained Weak Classifier Fusion for Early Detection of Alzheimer’s Disease

“…The size, the genders, the average ages and the average MMSE (Minimal Mental State Examination) are summarized in Table 1. These groups are similar to the ones used in [4,[7][8][9]. All 814 MRI were first segmented, normalized, modulated (correction of volume changes due to the normalization), and registered into a common space.…”

“…One part of these works focused on advanced machine learning techniques [6] while another part aimed to enhance the biomarker quality [4,7]. Among them, patch-based methods [8,9] demonstrated competitive AD prediction results. Despite these efforts, the current AD prognosis accuracy remains around 70%, that suggests the limitation of using (i) traditional features with advanced learning processes or (ii) high quality features with basic machine learning methods.…”

“…To this end, we first propose to extend the patch-based scoring method proposed in [8]. In this approach, the anatomical pattern similarity is estimated between the MCI test subject and two training populations (i.e., Cognitively Normal (CN) and AD) using a non-local patch-based scoring method.…”

“…For each voxel, a score (i.e., a grade) that measures the proximity to both training populations is computed. In [8], the a priori ROI-based strategy focused mainly on hippocampus and may discard other possible informative anatomical regions. To overcome this limitation, we propose to score the whole gray matter (GM).…”

“…Moreover, to be more robust to intensity normalization discrepancies between MRI, probabilities are used in place of intensities during patch comparison. Finally, while a local patch-based strategy is used in [9], a non-local approach is privileged to better handle inter-subject variability and registration error [8].…”