Andrea Chincarini

Standardized evaluation of algorithms for computer-aided diagnosis of dementia based on structural MRI: The CADDementia challenge

Algorithms for computer-aided diagnosis of dementia based on structural MRI have demonstrated high performance in the literature, but are difficult to compare as different data sets and methodology were used for evaluation. In addition, it is unclear how the algorithms would perform on previously unseen data, and thus, how they would perform in clinical practice when there is no real opportunity to adapt the algorithm to the data at hand. To address these comparability, generalizability and clinical applicability issues, we organized a grand challenge that aimed to objectively compare algorithms based on a clinically representative multi-center data set. Using clinical practice as the starting point, the goal was to reproduce the clinical diagnosis. Therefore, we evaluated algorithms for multi-class classification of three diagnostic groups: patients with probable Alzheimers disease, patients with mild cognitive impairment and healthy controls. The diagnosis based on clinical criteria was used as reference standard, as it was the best available reference despite its known limitations. For evaluation, a previously unseen test set was used consisting of 354 T1-weighted MRI scans with the diagnoses blinded. Fifteen research teams participated with a total of 29 algorithms. The algorithms were trained on a small training set (n=30) and optionally on data from other sources (e.g., the Alzheimers Disease Neuroimaging Initiative, the Australian Imaging Biomarkers and Lifestyle flagship study of aging). The best performing algorithm yielded an accuracy of 63.0% and an area under the receiver-operating-characteristic curve (AUC) of 78.8%. In general, the best performances were achieved using feature extraction based on voxel-based morphometry or a combination of features that included volume, cortical thickness, shape and intensity. The challenge is open for new submissions via the web-based framework: http://caddementia.grand-challenge.org.

Local MRI analysis approach in the diagnosis of early and prodromal Alzheimer's disease☆

BACKGROUND Medial temporal lobe (MTL) atrophy is one of the key biomarkers to detect early neurodegenerative changes in the course of Alzheimers disease (AD). There is active research aimed at identifying automated methodologies able to extract accurate classification indexes from T1-weighted magnetic resonance images (MRI). Such indexes should be fit for identifying AD patients as early as possible. SUBJECTS A reference group composed of 144AD patients and 189 age-matched controls was used to train and test the procedure. It was then applied on a study group composed of 302 MCI subjects, 136 having progressed to clinically probable AD (MCI-converters) and 166 having remained stable or recovered to normal condition after a 24month follow-up (MCI-non converters). All subjects came from the ADNI database. METHODS We sampled the brain with 7 relatively small volumes, mainly centered on the MTL, and 2 control regions. These volumes were filtered to give intensity and textural MRI-based features. Each filtered region was analyzed with a Random Forest (RF) classifier to extract relevant features, which were subsequently processed with a Support Vector Machine (SVM) classifier. Once a prediction model was trained and tested on the reference group, it was used to compute a classification index (CI) on the MCI cohort and to assess its accuracy in predicting AD conversion in MCI patients. The performance of the classification based on the features extracted by the whole 9 volumes is compared with that derived from each single volume. All experiments were performed using a bootstrap sampling estimation, and classifier performance was cross-validated with a 20-fold paradigm. RESULTS We identified a restricted set of image features correlated with the conversion to AD. It is shown that most information originate from a small subset of the total available features, and that it is enough to give a reliable assessment. We found multiple, highly localized image-based features which alone are responsible for the overall clinical diagnosis and prognosis. The classification index is able to discriminate Controls from AD with an Area Under Curve (AUC)=0.97 (sensitivity ≃89% at specificity ≃94%) and Controls from MCI-converters with an AUC=0.92 (sensitivity ≃89% at specificity ≃80%). MCI-converters are separated from MCI-non converters with AUC=0.74(sensitivity ≃72% at specificity ≃65%). FINDINGS The present automated MRI-based technique revealed a strong relationship between highly localized baseline-MRI features and the baseline clinical assessment. In addition, the classification index was also used to predict the probability of AD conversion within a time frame of two years. The definition of a single index combining local analysis of several regions can be useful to detect AD neurodegeneration in a typical MCI population.

Volume of interest-based [18F]fluorodeoxyglucose PET discriminates MCI converting to Alzheimer's disease from healthy controls. A European Alzheimer's Disease Consortium (EADC) study

An emerging issue in neuroimaging is to assess the diagnostic reliability of PET and its application in clinical practice. We aimed at assessing the accuracy of brain FDG-PET in discriminating patients with MCI due to Alzheimers disease and healthy controls. Sixty-two patients with amnestic MCI and 109 healthy subjects recruited in five centers of the European AD Consortium were enrolled. Group analysis was performed by SPM8 to confirm metabolic differences. Discriminant analyses were then carried out using the mean FDG uptake values normalized to the cerebellum computed in 45 anatomical volumes of interest (VOIs) in each hemisphere (90 VOIs) as defined in the Automated Anatomical Labeling (AAL) Atlas and on 12 meta-VOIs, bilaterally, obtained merging VOIs with similar anatomo-functional characteristics. Further, asymmetry indexes were calculated for both datasets. Accuracy of discrimination by a Support Vector Machine (SVM) and the AAL VOIs was tested against a validated method (PALZ). At the voxel level SMP8 showed a relative hypometabolism in the bilateral precuneus, and posterior cingulate, temporo-parietal and frontal cortices. Discriminant analysis classified subjects with an accuracy ranging between .91 and .83 as a function of data organization. The best values were obtained from a subset of 6 meta-VOIs plus 6 asymmetry values reaching an area under the ROC curve of .947, significantly larger than the one obtained by the PALZ score. High accuracy in discriminating MCI converters from healthy controls was reached by a non-linear classifier based on SVM applied on predefined anatomo-functional regions and inter-hemispheric asymmetries. Data pre-processing was automated and simplified by an in-house created Matlab-based script encouraging its routine clinical use. Further validation toward nonconverter MCI patients with adequately long follow-up is needed.

Automatic analysis of medial temporal lobe atrophy from structural MRIs for the early assessment of Alzheimer disease.

The purpose of this study is to develop a software for the extraction of the hippocampus and surrounding medial temporal lobe (MTL) regions from T1-weighted magnetic resonance (MR) images with no interactive input from the user, to introduce a novel statistical indicator, computed on the intensities in the automatically extracted MTL regions, which measures atrophy, and to evaluate the accuracy of the newly developed intensity-based measure of MTL atrophy to (a) distinguish between patients with Alzheimer disease (AD), patients with amnestic mild cognitive impairment (aMCI), and elderly controls by using established criteria for patients with AD and aMCI as the reference standard and (b) infer about the clinical outcome of aMCI patients. For the development of the software, the study included 61 patients with mild AD (17 men, 44 women; mean age±standard deviation (SD), 75.8years±7.8; Mini Mental State Examination (MMSE) score, 24.1±3.1), 42 patients with aMCI (11 men, 31 women; mean age±SD, 75.2years±4.9; MMSE score, 27.9±1.9), and 30 elderly healthy controls (10 men, 20 women; mean age±SD, 74.7years±5.2; MMSE score, 29.1±0.8). For the evaluation of the statistical indicator, 150 patients with mild AD (62 men, 88 women; mean age±SD, 76.3years±5.8; MMSE score, 23.2±4.1), 247 patients with aMCI (143 men, 104 women; mean age±SD, 75.3years±6.7; MMSE score, 27.0±1.8), and 135 elderly healthy controls (61 men, 74 women; mean age±SD, 76.4years±6.1). Fifty aMCI patients were evaluated every 6 months over a 3 year period to assess conversion to AD. For each participant, two subimages of the MTL regions were automatically extracted from T1-weighted MR images with high spatial resolution. An intensity-based MTL atrophy measure was found to separate control, MCI, and AD cohorts. Group differences wereassessed by using two-sample t test. Individual classification was analyzed by using receiver operating characteristic (ROC) curves. Compared to controls, significant differences in the intensity-based MTL atrophy measure were detected in both groups of patients (AD vs controls, 0.28±0.03 vs 0.34±0.03, P<0.001; aMCI vs controls, 0.31±0.03 vs 0.34±0.03, P<0.001). Moreover, the subgroup of aMCI converters was significantly different from controls (0.27±0.034 vs 0.34±0.03, P<0.001). Regarding the ROC curve for intergroup discrimination, the area under the curve was 0.863 for AD patients vs controls, 0.746 for all aMCI patients vs controls, and 0.880 for aMCI converters vs controls. With specificity set at 85%, the sensitivity was 74% for AD vs controls, 45% for aMCI vs controls, and 83% for aMCI converters vs controls. The automated analysis of MTL atrophy in the segmented volume is applied to the early assessment of AD, leading to the discrimination of aMCI converters with an average 3 year follow-up. This procedure can provide additional useful information in the early diagnosis of AD.

Integrating longitudinal information in hippocampal volume measurements for the early detection of Alzheimer's disease.

BACKGROUND Structural MRI measures for monitoring Alzheimers Disease (AD) progression are becoming instrumental in the clinical practice, and more so in the context of longitudinal studies. This investigation addresses the impact of four image analysis approaches on the longitudinal performance of the hippocampal volume. METHODS We present a hippocampal segmentation algorithm and validate it on a gold-standard manual tracing database. We segmented 460 subjects from ADNI, each subject having been scanned twice at baseline, 12-month and 24month follow-up scan (1.5T, T1 MRI). We used the bilateral hippocampal volume v and its variation, measured as the annualized volume change Λ=δv/year(mm(3)/y). Four processing approaches with different complexity are compared to maximize the longitudinal information, and they are tested for cohort discrimination ability. Reference cohorts are Controls vs. Alzheimers Disease (CTRL/AD) and CTRL vs. Mild Cognitive Impairment who subsequently progressed to AD dementia (CTRL/MCI-co). We discuss the conditions on v and the added value of Λ in discriminating subjects. RESULTS The age-corrected bilateral annualized atrophy rate (%/year) were: -1.6 (0.6) for CTRL, -2.2 (1.0) for MCI-nc, -3.2 (1.2) for MCI-co and -4.0 (1.5) for AD. Combined (v, Λ) discrimination ability gave an Area under the ROC curve (auc)=0.93 for CTRL vs AD and auc=0.88 for CTRL vs MCI-co. CONCLUSIONS Longitudinal volume measurements can provide meaningful clinical insight and added value with respect to the baseline provided the analysis procedure embeds the longitudinal information.

Hippocampal unified multi-atlas network (HUMAN): protocol and scale validation of a novel segmentation tool

In this study we present a novel fully automated Hippocampal Unified Multi-Atlas-Networks (HUMAN) algorithm for the segmentation of the hippocampus in structural magnetic resonance imaging. In multi-atlas approaches atlas selection is of crucial importance for the accuracy of the segmentation. Here we present an optimized method based on the definition of a small peri-hippocampal region to target the atlas learning with linear and non-linear embedded manifolds. All atlases were co-registered to a data driven template resulting in a computationally efficient method that requires only one test registration. The optimal atlases identified were used to train dedicated artificial neural networks whose labels were then propagated and fused to obtain the final segmentation. To quantify data heterogeneity and protocol inherent effects, HUMAN was tested on two independent data sets provided by the Alzheimers Disease Neuroimaging Initiative and the Open Access Series of Imaging Studies. HUMAN is accurate and achieves state-of-the-art performance (Dice[Formula: see text] and Dice[Formula: see text]). It is also a robust method that remains stable when applied to the whole hippocampus or to sub-regions (patches). HUMAN also compares favorably with a basic multi-atlas approach and a benchmark segmentation tool such as FreeSurfer.

Predictive Models Based on Support Vector Machines: Whole-Brain versus Regional Analysis of Structural MRI in the Alzheimer's Disease

Decision‐making systems trained on structural magnetic resonance imaging data of subjects affected by the Alzheimers disease (AD) and healthy controls (CTRL) are becoming widespread prognostic tools for subjects with mild cognitive impairment (MCI). This study compares the performances of three classification methods based on support vector machines (SVMs), using as initial sets of brain voxels (ie, features): (1) the segmented grey matter (GM); (2) regions of interest (ROIs) by voxel‐wise t‐test filtering; (3) parceled ROIs, according to prior knowledge. The recursive feature elimination (RFE) is applied in all cases to investigate whether feature reduction improves the classification accuracy. We analyzed more than 600 AD Neuroimaging Initiative (ADNI) subjects, training the SVMs on the AD/CTRL dataset, and evaluating them on a trial MCI dataset. The classification performance, evaluated as the area under the receiver operating characteristic curve (AUC), reaches AUC = (88.9 ± .5)% in 20‐fold cross‐validation on the AD/CTRL dataset, when the GM is classified as a whole. The highest discrimination accuracy between MCI converters and nonconverters is achieved when the SVM‐RFE is applied to the whole GM: with AUC reaching (70.7 ± .9)%, it outperforms both ROI‐based approaches in predicting the AD conversion.

Predicting the transition from normal aging to Alzheimer's disease: A statistical mechanistic evaluation of FDG-PET data.

The assessment of the degree of order of brain metabolism by means of a statistical mechanistic approach applied to FDG-PET, allowed us to characterize healthy subjects as well as patients with mild cognitive impairment and Alzheimers Disease (AD). The intensity signals from 24 volumes of interest were submitted to principal component analysis (PCA) giving rise to a major first principal component whose eigenvalue was a reliable cumulative index of order. This index linearly decreased from 77 to 44% going from normal aging to AD patients with intermediate conditions between these values (r=0.96, p<0.001). Bootstrap analysis confirmed the statistical significance of the results. The progressive detachment of different brain regions from the first component was assessed, allowing for a purely data driven reconstruction of already known maximally affected areas. We demonstrated for the first time the reliability of a single global index of order in discriminating groups of cognitively impaired patients with different clinical outcome. The second relevant finding was the identification of clusters of regions relevant to AD pathology progressively separating from the first principal component through different stages of cognitive impairment, including patients cognitively impaired but not converted to AD. This paved the way to the quantitative assessment of the functional networking status in individual patients.

Feature Selection Based on Machine Learning in MRIs for Hippocampal Segmentation

Neurodegenerative diseases are frequently associated with structural changes in the brain. Magnetic resonance imaging (MRI) scans can show these variations and therefore can be used as a supportive feature for a number of neurodegenerative diseases. The hippocampus has been known to be a biomarker for Alzheimer disease and other neurological and psychiatric diseases. However, it requires accurate, robust, and reproducible delineation of hippocampal structures. Fully automatic methods are usually the voxel based approach; for each voxel a number of local features were calculated. In this paper, we compared four different techniques for feature selection from a set of 315 features extracted for each voxel: (i) filter method based on the Kolmogorov-Smirnov test; two wrapper methods, respectively, (ii) sequential forward selection and (iii) sequential backward elimination; and (iv) embedded method based on the Random Forest Classifier on a set of 10 T1-weighted brain MRIs and tested on an independent set of 25 subjects. The resulting segmentations were compared with manual reference labelling. By using only 23 feature for each voxel (sequential backward elimination) we obtained comparable state-of-the-art performances with respect to the standard tool FreeSurfer.

Automatic temporal lobe atrophy assessment in prodromal AD: Data from the DESCRIPA study

In the framework of the clinical validation of research tools, this investigation presents a validation study of an automatic medial temporal lobe atrophy measure that is applied to a naturalistic population sampled from memory clinic patients across Europe.