Fully-automated approach to hippocampus segmentation using a graph-cuts algorithm combined with atlas-based segmentation and morphological opening

Clinical and Pathological Validation of CT-Based Regional Harmonization Methods of Amyloid PET

PURPOSE

The CT-based regional direct comparison Centiloid (dcCL) method was developed to harmonize and quantify regional β-amyloid (Aβ) burden. In the present study, we aimed to investigate correlations between the CT-based regional dcCL scales and Aβ pathological burdens and to validate the clinical utility using thresholds derived from pathological assessment.

PATIENTS AND METHODS

We included a pathological cohort of 63 cases and a clinical cohort of 4062 participants, and obtained modified Consortium to Establish a Registry for Alzheimer's Disease criteria (mCERAD) scores by assessment of neuritic plaque burdens in multiple areas of each cortical region. PET and CT images were processed using the CT-based regional dcCL method to calculate scales in 6 distinct regions.

RESULTS

The CT-based regional dcCL scales were correlated with neuritic plaque burdens represented by mCERAD scores, globally and regionally ( r = 0.56~0.76). In addition, striatum dcCL scales reflected Aβ involvement in the striatum ( P < 0.001). The regional dcCL scales could predict significant Aβ deposition in specific brain regions with high accuracy: area under the receiver operating characteristic curve of 0.81-0.97 with an mCERAD cutoff of 1.5 and area under the receiver operating characteristic curve of 0.88-0.93 with an mCERAD cutoff of 0.5. When applying the dcCL thresholds of 1.5 mCERAD scores, the G(-)R(+) group showed lower performances in memory and global cognitive functions and had less hippocampal volume compared with the G(-)R(-) group ( P < 0.001). However, when applying the dcCL thresholds of 0.5 mCERAD scores, there were no differences in the global cognitive functions between the 2 groups.

CONCLUSIONS

The thresholds of regional dcCL scales derived from pathological assessments might provide clinicians with a better understanding of biomarker-guided diagnosis and distinguishable clinical phenotypes, which are particularly useful when harmonizing different PET ligands with only PET/CT.

Distinct effects of cholesterol profile components on amyloid and vascular burdens

BACKGROUND

Cholesterol plays important roles in β-amyloid (Aβ) metabolism and atherosclerosis. However, the relationships of plasma cholesterol levels with Aβ and cerebral small vessel disease (CSVD) burdens are not fully understood in Asians. Herein, we investigated the relationships between plasma cholesterol profile components and Aβ and CSVD burdens in a large, non-demented Korean cohort.

METHODS

We enrolled 1,175 non-demented participants (456 with unimpaired cognition [CU] and 719 with mild cognitive impairment [MCI]) aged ≥ 45 years who underwent Aβ PET at the Samsung Medical Center in Korea. We performed linear regression analyses with each cholesterol (low-density lipoprotein cholesterol [LDL-c], high-density lipoprotein cholesterol [HDL-c], and triglyceride) level as a predictor and each image marker (Aβ uptake on PET, white matter hyperintensity [WMH] volume, and hippocampal volume) as an outcome after controlling for potential confounders.

RESULTS

Increased LDL-c levels (β = 0.014 to 0.115, p = 0.013) were associated with greater Aβ uptake, independent of the APOE e4 allele genotype and lipid-lowering medication. Decreased HDL-c levels (β =  - 0.133 to - 0.006, p = 0.032) were predictive of higher WMH volumes. Increased LDL-c levels were also associated with decreased hippocampal volume (direct effect β =  - 0.053, p = 0.040), which was partially mediated by Aβ uptake (indirect effect β =  - 0.018, p = 0.006).

CONCLUSIONS

Our findings highlight that increased LDL-c and decreased HDL-c levels are important risk factors for Aβ and CSVD burdens, respectively. Furthermore, considering that plasma cholesterol profile components are potentially modified by diet, exercise, and pharmacological agents, our results provide evidence that regulating LDL-c and HDL-c levels is a potential strategy to prevent dementia.

Clinical effects of novel susceptibility genes for beta-amyloid: a gene-based association study in the Korean population

Amyloid-beta (Aβ) is a pathological hallmark of Alzheimer's disease (AD). We aimed to identify genes related to Aβ uptake in the Korean population and investigate the effects of these novel genes on clinical outcomes, including neurodegeneration and cognitive impairments. We recruited a total of 759 Korean participants who underwent neuropsychological tests, brain magnetic resonance imaging, 18F-flutemetamol positron emission tomography, and microarray genotyping data. We performed gene-based association analysis, and also performed expression quantitative trait loci and network analysis. In genome-wide association studies, no single nucleotide polymorphism (SNP) passed the genome-wide significance threshold. In gene-based association analysis, six genes (LCMT1, SCRN2, LRRC46, MRPL10, SP6, and OSBPL7) were significantly associated with Aβ standardised uptake value ratio in the brain. The three most significant SNPs (rs4787307, rs9903904, and rs11079797) on these genes are associated with the regulation of the LCMT1, OSBPL7, and SCRN2 genes, respectively. These SNPs are involved in decreasing hippocampal volume and cognitive scores by mediating Aβ uptake. The 19 enriched gene sets identified by pathway analysis included axon and chemokine activity. Our findings suggest novel susceptibility genes associated with the uptake of Aβ, which in turn leads to worse clinical outcomes. Our findings might lead to the discovery of new AD treatment targets.

Classification and prediction of cognitive trajectories of cognitively unimpaired individuals

Objectives

Efforts to prevent Alzheimer's disease (AD) would benefit from identifying cognitively unimpaired (CU) individuals who are liable to progress to cognitive impairment. Therefore, we aimed to develop a model to predict cognitive decline among CU individuals in two independent cohorts.

Methods

A total of 407 CU individuals from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and 285 CU individuals from the Samsung Medical Center (SMC) were recruited in this study. We assessed cognitive outcomes by using neuropsychological composite scores in the ADNI and SMC cohorts. We performed latent growth mixture modeling and developed the predictive model.

Results

Growth mixture modeling identified 13.8 and 13.0% of CU individuals in the ADNI and SMC cohorts, respectively, as the "declining group." In the ADNI cohort, multivariable logistic regression modeling showed that increased amyloid-β (Aβ) uptake (β [SE]: 4.852 [0.862], p < 0.001), low baseline cognitive composite scores (β [SE]: -0.274 [0.070], p < 0.001), and reduced hippocampal volume (β [SE]: -0.952 [0.302], p = 0.002) were predictive of cognitive decline. In the SMC cohort, increased Aβ uptake (β [SE]: 2.007 [0.549], p < 0.001) and low baseline cognitive composite scores (β [SE]: -4.464 [0.758], p < 0.001) predicted cognitive decline. Finally, predictive models of cognitive decline showed good to excellent discrimination and calibration capabilities (C-statistic = 0.85 for the ADNI model and 0.94 for the SMC model).

Conclusion

Our study provides novel insights into the cognitive trajectories of CU individuals. Furthermore, the predictive model can facilitate the classification of CU individuals in future primary prevention trials.

Clinical outcomes of increased focal amyloid uptake in individuals with subthreshold global amyloid levels

Background

Although the standardized uptake value ratio (SUVR) method is objective and simple, cut-off optimization using global SUVR values may not reflect focal increased uptake in the cerebrum. The present study investigated clinical and neuroimaging characteristics according to focally increased β-amyloid (Aβ) uptake and global Aβ status.

Methods

We recruited 968 participants with cognitive continuum. All participants underwent neuropsychological tests and 498 ¹⁸F-florbetaben (FBB) amyloid positron emission tomography (PET) and 470 ¹⁸F-flutemetamol (FMM) PET. Each PET scan was assessed in 10 regions (left and right frontal, lateral temporal, parietal, cingulate, and striatum) with focal-quantitative SUVR-based cutoff values for each region by using an iterative outlier approach.

Results

A total of 62 (6.4%) subjects showed increased focal Aβ uptake with subthreshold global Aβ status [global (-) and focal (+) Aβ group, G(-)F(+) group]. The G(-)F(+) group showed worse performance in memory impairment (p < 0.001), global cognition (p = 0.009), greater hippocampal atrophy (p = 0.045), compared to those in the G(-)F(-). Participants with widespread Aβ involvement in the whole region [G(+)] showed worse neuropsychological (p < 0.001) and neuroimaging features (p < 0.001) than those with focal Aβ involvement G(-)F(+).

Conclusion

Our findings suggest that individuals show distinctive clinical outcomes according to focally increased Aβ uptake and global Aβ status. Thus, researchers and clinicians should pay more attention to focal increased Aβ uptake in addition to global Aβ status.

Distinct volumetric features of cerebrospinal fluid distribution in idiopathic normal-pressure hydrocephalus and Alzheimer's disease

Objective

The aims of the study were to measure the cerebrospinal fluid (CSF) volumes in the lateral ventricle, high-convexity subarachnoid space, and Sylvian fissure region in patients with idiopathic normal-pressure hydrocephalus (INPH) and Alzheimer’s disease (AD), and to evaluate differences in these volumes between INPH and AD groups and healthy controls.

Methods

Forty-nine INPH patients, 59 AD patients, and 26 healthy controls were imaged with automated three-dimensional volumetric MRI.

Results

INPH patients had larger lateral ventricles and CSF spaces of the Sylvian fissure region and smaller high-convexity subarachnoid spaces than other groups, and AD patients had larger lateral ventricles and CSF spaces of the Sylvian fissure region than the control group. The INPH group showed a negative correlation between lateral ventricle and high-convexity subarachnoid space volumes, while the AD group showed a positive correlation between lateral ventricle volume and volume for CSF spaces of the Sylvian fissure region. The ratio of lateral ventricle to high-convexity subarachnoid space volumes yielded an area under the curve of 0.990, differentiating INPH from AD.

Conclusions

Associations between CSF volumes suggest that there might be different mechanisms between INPH and AD to explain their respective lateral ventricular dilations. The ratio of lateral ventricle to high-convexity subarachnoid space volumes distinguishes INPH from AD with good diagnostic sensitivity and specificity. We propose to refer to this ratio as the VOSS (ventricle over subarachnoid space) index.

Longitudinal Amyloid Cognitive Composite in Preclinical Alzheimer's Disease

BACKGROUND

Previous studies have developed several cognitive composites in preclinical AD. However, more sensitive measures to track cognitive changes and therapeutic efficacy in preclinical Alzheimer's disease (AD) are needed considering diverse sociocultural and linguistic backgrounds. This study developed a composite score that can sensitively detect the Aβ-related cognitive trajectory of preclinical AD using Korean data.

METHODS

A total of 196 cognitively normal (CN) participants who underwent amyloid positron emission tomography were followed-up with neuropsychological assessments. We developed the Longitudinal Amyloid cognitive Composite in Preclinical AD (LACPA) using the linear mixed-effects model (LMM) and z-scores. The LMM was also used to investigate the longitudinal sensitivity of LACPA and the association between time-varying brain atrophy and LACPA.

RESULTS

Considering the group-time interaction effects of each subtest, the Seoul Verbal Learning Test-Elderly's version (SVLT-E) immediate recall/delayed recall/recognition, the Korean Trail Making Test B time, and the Korean Mini-Mental State Examination were selected as components of LACPA. LACPA exhibited a significant group-time interaction effect between the Aβ+ and Aβ- groups (t = -3.288, p = 0.001). Associations between time-varying LACPA and brain atrophy were found in the bilateral medial temporal, right lateral parietal, and right lateral frontal regions, and hippocampal volume.

CONCLUSION

LACPA may contribute to reduction in time and financial burden when monitoring Aβ-related cognitive decline and therapeutic efficacy of the disease-modifying agents specifically targeting Aβ in secondary prevention trials.

Hippocampus Segmentation Method Based on Subspace Patch-Sparsity Clustering in Noisy Brain MRI

Since the hippocampus is of small size, low contrast, and irregular shape, a novel hippocampus segmentation method based on subspace patch-sparsity clustering in brain MRI is proposed to improve the segmentation accuracy, which requires that the representation coefficients in different subspaces should be as sparse as possible, while the representation coefficients in the same subspace should be as average as possible. By restraining the coefficient matrix with the patch-sparse constraint, the coefficient matrix contains a patch-sparse structure, which is helpful to the hippocampus segmentation. The experimental results show that our proposed method is effective in the noisy brain MRI data, which can well deal with hippocampus segmentation problem.

Prediction of tau accumulation in prodromal Alzheimer's disease using an ensemble machine learning approach

We developed machine learning (ML) algorithms to predict abnormal tau accumulation among patients with prodromal AD. We recruited 64 patients with prodromal AD using the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset. Supervised ML approaches based on the random forest (RF) and a gradient boosting machine (GBM) were used. The GBM resulted in an AUC of 0.61 (95% confidence interval [CI] 0.579–0.647) with clinical data (age, sex, years of education) and a higher AUC of 0.817 (95% CI 0.804–0.830) with clinical and neuropsychological data. The highest AUC was 0.86 (95% CI 0.839–0.885) achieved with additional information such as cortical thickness in clinical data and neuropsychological results. Through the analysis of the impact order of the variables in each ML classifier, cortical thickness of the parietal lobe and occipital lobe and neuropsychological tests of memory domain were found to be more important features for each classifier. Our ML algorithms predicting tau burden may provide important information for the recruitment of participants in potential clinical trials of tau targeting therapies.

Staging and quantification of florbetaben PET images using machine learning: impact of predicted regional cortical tracer uptake and amyloid stage on clinical outcomes

Purpose

We developed a machine learning–based classifier for in vivo amyloid positron emission tomography (PET) staging, quantified cortical uptake of the PET tracer by using a machine learning method, and investigated the impact of these amyloid PET parameters on clinical and structural outcomes.

Methods

A total of 337 ¹⁸F-florbetaben PET scans obtained at Samsung Medical Center were assessed. We defined a feature vector representing the change in PET tracer uptake from grey to white matter. Using support vector machine (SVM) regression and SVM classification, we quantified the cortical uptake as predicted regional cortical tracer uptake (pRCTU) and categorised the scans as positive and negative. Positive scans were further classified into two stages according to the striatal uptake. We compared outcome parameters among stages and further assessed the association between the pRCTU and outcome variables. Finally, we performed path analysis to determine mediation effects between PET variables.

Results

The classification accuracy was 97.3% for cortical amyloid positivity and 91.1% for striatal positivity. The left frontal and precuneus/posterior cingulate regions, as well as the anterior portion of the striatum, were important in determination of stages. The clinical scores and magnetic resonance imaging parameters showed negative associations with PET stage. However, except for the hippocampal volume, most outcomes were associated with the stage through the complete mediation effect of pRCTU.

Conclusion

Using a machine learning algorithm, we achieved high accuracy for in vivo amyloid PET staging. The in vivo amyloid stage was associated with cognitive function and cerebral atrophy mostly through the mediation effect of cortical amyloid.

Electronic supplementary material

The online version of this article (10.1007/s00259-019-04663-3) contains supplementary material, which is available to authorized users.

Multi-Modality Sparse Representation for Alzheimer's Disease Classification

BACKGROUND

Alzheimer's disease (AD) and mild cognitive impairment (MCI) are age-related neurodegenerative diseases characterized by progressive loss of memory and irreversible cognitive functions. The hippocampus, a brain area critical for learning and memory processes, is especially susceptible to damage at early stages of AD.

OBJECTIVE

We aimed to develop prediction model using a multi-modality sparse representation approach.

METHODS

We proposed a sparse representation approach to the hippocampus using structural T1-weighted magnetic resonance imaging (MRI) and 18-fluorodeoxyglucose-positron emission tomography (FDG-PET) to distinguish AD/MCI from healthy control subjects (HCs). We considered structural and function information for the hippocampus and applied a sparse patch-based approach to effectively reduce the dimensions of neuroimaging biomarkers.

RESULTS

In experiments using Alzheimer's Disease Neuroimaging Initiative data, our proposed method demonstrated more reliable than previous classification studies. The effects of different parameters on segmentation accuracy were also evaluated. The mean classification accuracy obtained with our proposed method was 0.94 for AD/HCs, 0.82 for MCI/HCs, and 0.86 for AD/MCI.

CONCLUSION

We extracted multi-modal features from automatically defined hippocampal regions of training subjects and found this method to be discriminative and robust for AD and MCI classification. The extraction of features in T1 and FDG-PET images is expected to improve classification performance due to the relationship between brain structure and function.

Testing a deep convolutional neural network for automated hippocampus segmentation in a longitudinal sample of healthy participants

Subtle changes in hippocampal volumes may occur during both physiological and pathophysiological processes in the human brain. Assessing hippocampal volumes manually is a time-consuming procedure, however, creating a need for automated segmentation methods that are both fast and reliable over time. Segmentation algorithms that employ deep convolutional neural networks (CNN) have emerged as a promising solution for large longitudinal neuroimaging studies. However, for these novel algorithms to be useful in clinical studies, the accuracy and reproducibility should be established on independent datasets. Here, we evaluate the performance of a CNN-based hippocampal segmentation algorithm that was developed by Thyreau and colleagues - Hippodeep. We compared its segmentation outputs to manual segmentation and FreeSurfer 6.0 in a sample of 200 healthy participants scanned repeatedly at seven sites across Canada, as part of the Canadian Biomarker Integration Network in Depression consortium. The algorithm demonstrated high levels of stability and reproducibility of volumetric measures across all time points compared to the other two techniques. Although more rigorous testing in clinical populations is necessary, this approach holds promise as a viable option for tracking volumetric changes in longitudinal neuroimaging studies.

Amyloid Deposits and Idiopathic Normal-Pressure Hydrocephalus: An 18F-Florbetaben Study

Background: The first aim of our study was to determine whether cortical 18F-florbetaben retention was different between healthy controls and idiopathic normal-pressure hydrocephalus (INPH) patients. Our second aim was to investigate whether there were any relationships between 18F-florbetaben retention and either hippocampal volume or clinical symptoms in INPH patients. Methods: Seventeen patients diagnosed with INPH and 8 healthy controls underwent studies with magnetic resonance imaging and 18F-florbetaben positron emission tomography imaging. Results: Automated region-of-interest analysis showed significant increases in 18F-florbetaben uptake in several brain regions in INPH patients compared to control subjects, with especially remarkable increases in the frontal (bilateral), parietal (bilateral), and occipital (bilateral) cortices. In the INPH group, right hippocampal volume was found to be negatively correlated with right frontal 18F-florbetaben retention. Korean-Mini Mental State Examination scores negatively correlated with right occipital 18F-florbetaben retention. Higher 18F-florbetaben retention correlated significantly with a higher Clinical Dementia Rating Scale score in the right occipital cortex. Conclusions: Our results indicate that INPH might be a disease exhibiting a characteristic pattern of cortical 18F-florbetaben retention. 18F-florbetaben retention in the frontal cortex may be related to hippocampal neuronal degeneration. Our findings may also help us understand the potential pathophysiology of cognitive impairments associated with INPH.

Identification of ghost artifact using texture analysis in pediatric spinal cord diffusion tensor images

PURPOSE

Ghost artifacts are a major contributor to degradation of spinal cord diffusion tensor images. A multi-stage post-processing pipeline was designed, implemented and validated to automatically remove ghost artifacts arising from reduced field of view diffusion tensor imaging (DTI) of the pediatric spinal cord.

METHOD

A total of 12 pediatric subjects including 7 healthy subjects (mean age=11.34years) with no evidence of spinal cord injury or pathology and 5 patients (mean age=10.96years) with cervical spinal cord injury were studied. Ghost/true cords, labeled as region of interests (ROIs), in non-diffusion weighted b0 images were segmented automatically using mathematical morphological processing. Initially, 21 texture features were extracted from each segmented ROI including 5 first-order features based on the histogram of the image (mean, variance, skewness, kurtosis and entropy) and 16s-order feature vector elements, incorporating four statistical measures (contrast, correlation, homogeneity and energy) calculated from co-occurrence matrices in directions of 0°, 45°, 90° and 135°. Next, ten features with a high value of mutual information (MI) relative to the pre-defined target class and within the features were selected as final features which were input to a trained classifier (adaptive neuro-fuzzy interface system) to separate the true cord from the ghost cord.

RESULTS

The implemented pipeline was successfully able to separate the ghost artifacts from true cord structures. The results obtained from the classifier showed a sensitivity of 91%, specificity of 79%, and accuracy of 84% in separating the true cord from ghost artifacts.

CONCLUSION

The results show that the proposed method is promising for the automatic detection of ghost cords present in DTI images of the spinal cord. This step is crucial towards development of accurate, automatic DTI spinal cord post processing pipelines.

The bumps under the hippocampus

Shown in every neuroanatomy textbook, a key morphological feature is the bumpy ridges, which we refer to as hippocampal dentation, on the inferior aspect of the hippocampus. Like the folding of the cerebral cortex, hippocampal dentation allows for greater surface area in a confined space. However, examining numerous approaches to hippocampal segmentation and morphology analysis, virtually all published 3D renderings of the hippocampus show the inferior surface to be quite smooth or mildly irregular; we have rarely seen the characteristic bumpy structure on reconstructed 3D surfaces. The only exception is a 9.4T postmortem study (Yushkevich et al. [2009]: NeuroImage 44:385-398). An apparent question is, does this indicate that this specific morphological signature can only be captured using ultra high-resolution techniques? Or, is such information buried in the data we commonly acquire, awaiting a computation technique that can extract and render it clearly? In this study, we propose an automatic and robust super-resolution technique that captures the fine scale morphometric features of the hippocampus based on common 3T MR images. The method is validated on 9.4T ultra-high field images and then applied on 3T data sets. This method opens possibilities of future research on the hippocampus and other sub-cortical structural morphometry correlating the degree of dentation with a range of diseases including epilepsy, Alzheimer's disease, and schizophrenia. Hum Brain Mapp 39:472-490, 2018. © 2017 Wiley Periodicals, Inc.

Hippocampus Segmentation Based on Local Linear Mapping

We propose local linear mapping (LLM), a novel fusion framework for distance field (DF) to perform automatic hippocampus segmentation. A k-means cluster method is propose for constructing magnetic resonance (MR) and DF dictionaries. In LLM, we assume that the MR and DF samples are located on two nonlinear manifolds and the mapping from the MR manifold to the DF manifold is differentiable and locally linear. We combine the MR dictionary using local linear representation to present the test sample, and combine the DF dictionary using the corresponding coefficients derived from local linear representation procedure to predict the DF of the test sample. We then merge the overlapped predicted DF patch to obtain the DF value of each point in the test image via a confidence-based weighted average method. This approach enabled us to estimate the label of the test image according to the predicted DF. The proposed method was evaluated on brain images of 35 subjects obtained from SATA dataset. Results indicate the effectiveness of the proposed method, which yields mean Dice similarity coefficients of 0.8697, 0.8770 and 0.8734 for the left, right and bi-lateral hippocampus, respectively.

Level Set Based Hippocampus Segmentation in MR Images with Improved Initialization Using Region Growing

The hippocampus has been known as one of the most important structures referred to as Alzheimer's disease and other neurological disorders. However, segmentation of the hippocampus from MR images is still a challenging task due to its small size, complex shape, low contrast, and discontinuous boundaries. For the accurate and efficient detection of the hippocampus, a new image segmentation method based on adaptive region growing and level set algorithm is proposed. Firstly, adaptive region growing and morphological operations are performed in the target regions and its output is used for the initial contour of level set evolution method. Then, an improved edge-based level set method utilizing global Gaussian distributions with different means and variances is developed to implement the accurate segmentation. Finally, gradient descent method is adopted to get the minimization of the energy equation. As proved by experiment results, the proposed method can ideally extract the contours of the hippocampus that are very close to manual segmentation drawn by specialists.

Comparative performance evaluation of automated segmentation methods of hippocampus from magnetic resonance images of temporal lobe epilepsy patients

PURPOSE

Segmentation of the hippocampus from magnetic resonance (MR) images is a key task in the evaluation of mesial temporal lobe epilepsy (mTLE) patients. Several automated algorithms have been proposed although manual segmentation remains the benchmark. Choosing a reliable algorithm is problematic since structural definition pertaining to multiple edges, missing and fuzzy boundaries, and shape changes varies among mTLE subjects. Lack of statistical references and guidance for quantifying the reliability and reproducibility of automated techniques has further detracted from automated approaches. The purpose of this study was to develop a systematic and statistical approach using a large dataset for the evaluation of automated methods and establish a method that would achieve results better approximating those attained by manual tracing in the epileptogenic hippocampus.

METHODS

A template database of 195 (81 males, 114 females; age range 32-67 yr, mean 49.16 yr) MR images of mTLE patients was used in this study. Hippocampal segmentation was accomplished manually and by two well-known tools (FreeSurfer and hammer) and two previously published methods developed at their institution [Automatic brain structure segmentation (ABSS) and LocalInfo]. To establish which method was better performing for mTLE cases, several voxel-based, distance-based, and volume-based performance metrics were considered. Statistical validations of the results using automated techniques were compared with the results of benchmark manual segmentation. Extracted metrics were analyzed to find the method that provided a more similar result relative to the benchmark.

RESULTS

Among the four automated methods, ABSS generated the most accurate results. For this method, the Dice coefficient was 5.13%, 14.10%, and 16.67% higher, Hausdorff was 22.65%, 86.73%, and 69.58% lower, precision was 4.94%, -4.94%, and 12.35% higher, and the root mean square (RMS) was 19.05%, 61.90%, and 65.08% lower than LocalInfo, FreeSurfer, and hammer, respectively. The Bland-Altman similarity analysis revealed a low bias for the ABSS and LocalInfo techniques compared to the others.

CONCLUSIONS

The ABSS method for automated hippocampal segmentation outperformed other methods, best approximating what could be achieved by manual tracing. This study also shows that four categories of input data can cause automated segmentation methods to fail. They include incomplete studies, artifact, low signal-to-noise ratio, and inhomogeneity. Different scanner platforms and pulse sequences were considered as means by which to improve reliability of the automated methods. Other modifications were specially devised to enhance a particular method assessed in this study.

Multimodal Discrimination of Alzheimer's Disease Based on Regional Cortical Atrophy and Hypometabolism

Structural MR image (MRI) and ¹⁸F-Fluorodeoxyglucose-positron emission tomography (FDG-PET) have been widely employed in diagnosis of both Alzheimer’s disease (AD) and mild cognitive impairment (MCI) pathology, which has led to the development of methods to distinguish AD and MCI from normal controls (NC). Synaptic dysfunction leads to a reduction in the rate of metabolism of glucose in the brain and is thought to represent AD progression. FDG-PET has the unique ability to estimate glucose metabolism, providing information on the distribution of hypometabolism. In addition, patients with AD exhibit significant neuronal loss in cerebral regions, and previous AD research has shown that structural MRI can be used to sensitively measure cortical atrophy. In this paper, we introduced a new method to discriminate AD from NC based on complementary information obtained by FDG and MRI. For accurate classification, surface-based features were employed and 12 predefined regions were selected from previous studies based on both MRI and FDG-PET. Partial least square linear discriminant analysis was employed for making diagnoses. We obtained 93.6% classification accuracy, 90.1% sensitivity, and 96.5% specificity in discriminating AD from NC. The classification scheme had an accuracy of 76.5% and sensitivity and specificity of 46.5% and 89.6%, respectively, for discriminating MCI from AD. Our method exhibited a superior classification performance compared with single modal approaches and yielded parallel accuracy to previous multimodal classification studies using MRI and FDG-PET.