Evidence for a differential contribution of early perceptual and late cognitive processes during encoding to episodic memory impairment in schizophrenia.

OBJECTIVES: Schizophrenia is characterised by significant episodic memory impairment that is thought to be related to problems with encoding, however the neuro-functional mechanisms underlying these deficits are not well understood. The present study used a subsequent recognition memory paradigm and event-related potentials (ERPs) to investigate temporal aspects of episodic memory encoding deficits in schizophrenia. METHODS: Electroencephalographic data was recorded in 24 patients and 19 healthy controls whilst participants categorised single words as pleasant/unpleasant. ERPs were generated to subsequently recognised versus unrecognised words on the basis of a forced-choice recognition memory task. Subsequent memory effects were examined with the late positive component (LPP). Group differences in N1, P2, N400 and LPP were examined for words correctly recognised. RESULTS: Patients performed more poorly than controls on the recognition task. During encoding patients had significantly reduced N400 and LPP amplitudes than controls. LPP amplitude correlated with task performance however amplitudes did not differ between patients and controls as a function of subsequent memory. No significant differences in N1 or P2 amplitude or latency were observed. CONCLUSIONS: The present results indicate that early sensory processes are intact and dysfunctional higher order cognitive processes during encoding are contributing to episodic memory impairments in schizophrenia.

White matter correlates of episodic memory encoding and retrieval in schizophrenia.

Episodic memory (EM) impairments in schizophrenia (SZ) are predictive of functional outcome and are a potential endophenotype of the disorder. The current study investigated the neuroanatomical correlates of EM encoding and retrieval in SZ with structural magnetic resonance and diffusion tensor imaging (DTI) measures in 22 patients with SZ and 22 age- and gender-matched healthy controls. Tract-based Spatial Statistics (TBSS) was used to investigate microstructural alterations in white matter (WM), while FreeSurfer surface-based analysis was used to determine abnormalities in grey matter (GM) and WM volumetrics and cortical thickness. Compared to controls, patients demonstrated GM deficits in temporal and parietal regions and lower fractional anisotropy (FA) of WM in diffuse brain regions. Patients also demonstrated reduced functioning in both encoding and retention of auditory-verbal EM. Among patients but not controls, EM encoding correlated with WM volume in the orbitofrontal cortex and increased radial diffusivity in the fornix, whereas EM retrieval correlated with WM volume in posterior parietal cortex. These findings suggest a differential role for frontal and parietal WM in EM encoding and retrieval processes, while myelin integrity of the fornix may play a specific role in mediating EM encoding processes in SZ.

The relevance of attention in schizophrenia P50 paired stimulus studies.

OBJECTIVE: P50 suppression refers to the P50 ERP amplitude-reduction to the second (S2) relative to the first (S1) of identical brief auditory stimuli (SOA=500ms). Its reduction in schizophrenia is argued to represent impaired inhibitory input (II) mechanisms. Enhancing attention enhances II functionality (reducing S2P50 amplitude and increasing P50 difference) in healthy subjects. We determined whether the effect of attention on P50 suppression differs between schizophrenia patients (SCZ) and controls (CON) and thus is a confound in P50 schizophrenia research. METHODS: We manipulated the direction of attention (attention, non-attention) in 21 SCZ and 18 CON in the P50 suppression task. RESULTS: Directing attention towards stimulus pairs (versus non-attention) increased P50 suppression (P50 difference). This effect differed between groups, with attention increasing S1P50, reducing S2P50 and increasing P50 suppression (P50 difference and reducing P50 ratio) in CON only. No group differences were found for P50 difference or ratio. CONCLUSIONS: Attention is a confound in schizophrenia P50 research and thus should be carefully controlled. When attention was controlled, P50 group differences were not found. SIGNIFICANCE: The SCZ-CON P50 difference reported in the literature may be related to uncontrolled attention (and not impaired P50 suppression per se).

Age effects on cortical thickness in cognitively normal elderly individuals.

BACKGROUND/AIMS: Atrophy in both grey and white matter is found in normal aging. The prefrontal cortex and the frontal lobe white matter are thought to be the most affected regions. Our aim was to examine the effects of normal aging on cortical grey matter using a 3D quantitative cortical mapping method. METHODS: We analyzed 1.5-tesla brain magnetic resonance imaging data from 44 cognitively normal elderly subjects using cortical pattern matching and cortical thickness analyses. Linear regression analysis was used to study the effect of age on cortical thickness. 3D map-wide correction for multiple comparisons was conducted with permutation analyses using a threshold of p < 0.01. RESULTS: We found a significant negative association between age and cortical thickness in the right hemisphere (pcorrected = 0.009) and a trend level association in the left hemisphere (pcorrected = 0.081). Age-related changes were greatest in the sensorimotor, bilateral dorsal anterior cingulate and supplementary motor cortices, and the right posterior middle and inferior frontal gyri. Age effects greater in the medial than lateral visual association cortices were also seen bilaterally. CONCLUSION: Our novel method further validates that normal aging results in diffuse cortical thinning that is most pronounced in the frontal and visual association cortices.

Mapping cortical atrophy in Parkinson's disease patients with dementia.

BACKGROUND: Cognitive impairment is very common in patients with Parkinson's disease (PD). Brain changes accompanying cognitive decline in PD are still not fully established. METHODS: We applied cortical pattern matching and cortical thickness analyses to the three-dimensional T1-weighted brain MRI scans of 14 age-matched cognitively normal elderly (NC), 12 cognitively normal PD (PDC), and 11 PD dementia (PDD) subjects. We used linear regression models to investigate the effect of diagnosis on cortical thickness. All maps were adjusted for multiple comparisons using permutation testing with a threshold p < 0.01. RESULTS: PDD showed significantly thinner bilateral sensorimotor, perisylvian, lateral parietal, as well as right posterior cingulate, parieto-occipital, inferior temporal and lateral frontal cortices relative to NC (left p(corrected) = 0.06, right p(corrected) = 0.009). PDD showed significantly thinner bilateral sensorimotor, right frontal and right parietal-occipital cortices relative to PDC (right p(corrected) = 0.05). The absolute difference in cortical thickness between PDD and the other diagnostic groups ranged from 3% to 19%. CONCLUSION: Our data shows that cognitive decline in PD is associated with cortical atrophy. PDD subjects have the most widespread gray matter atrophy suggesting more cortical involvement as PD patients progress to dementia.

Ventricular enlargement and its clinical correlates in the imaging cohort from the ADCS MCI donepezil/vitamin E study.

We analyzed the baseline and 3-year T1-weighted magnetic resonance imaging data of 110 amnestic mild cognitive impairment (MCI) participants with minimal hippocampal atrophy at baseline from the Alzheimer's Disease Cooperative Study group MCI Donepezil/Vitamin E trial. Forty-six subjects converted to Alzheimer disease (AD) (MCIc), whereas 64 remained stable (MCInc). We used the radial distance technique to examine the differences in lateral ventricle shape and size between MCIc and MCInc and the associations between ventricular enlargement and cognitive decline. MCIc group had significantly larger frontal and right body/occipital horns relative to MCInc at baseline and significantly larger bilateral frontal, body/occipital, and left temporal horns at follow-up. Global cognitive decline measured with AD Assessment scale cognitive subscale and Mini-Mental State Examination and decline in activities of daily living (ADL) were associated with posterior lateral ventricle enlargement. Decline in AD Assessment scale cognitive subscale and ADL were associated with left temporal and decline in Mini-Mental State Examination with right temporal horn enlargement. After correction for baseline hippocampal volume, decline in ADL showed a significant association with right frontal horn enlargement. Executive decline was associated with right frontal and left temporal horn enlargement.

Hippocampal atrophy and ventricular enlargement in normal aging, mild cognitive impairment (MCI), and Alzheimer Disease.

Alzheimer disease (AD) is the most common type of dementia worldwide. Hippocampal atrophy and ventricular enlargement have been associated with AD but also with normal aging. We analyzed 1.5-T brain magnetic resonance imaging data from 46 cognitively normal elderly individuals (NC), 33 mild cognitive impairment and 43 AD patients. Hippocampal and ventricular analyses were conducted with 2 novel semiautomated segmentation approaches followed by the radial distance mapping technique. Multiple linear regression was used to assess the effects of age and diagnosis on hippocampal and ventricular volumes and radial distance. In addition, 3-dimensional map correction for multiple comparisons was made with permutation testing. As expected, most significant hippocampal atrophy and ventricular enlargement were seen in the AD versus NC comparison. Mild cognitive impairment patients showed intermediate levels of hippocampal atrophy and ventricular enlargement. Significant effects of age on hippocampal volume and radial distance were seen in the pooled sample and in the NC and AD groups considered separately. Age-associated differences were detected in all hippocampal subfields and in the frontal and body/occipital horn portions of the lateral ventricles. Aging affects both the hippocampus and lateral ventricles independent of AD pathology, and should be included as covariate in all structural, hippocampal, and ventricular analyses when possible.

Effects of ApoE4 and maternal history of dementia on hippocampal atrophy.

We applied an automated hippocampal segmentation technique based on adaptive boosting (AdaBoost) to the 1.5 T magnetic resonance imaging (MRI) baseline and 1-year follow-up data of 243 subjects with mild cognitive impairment (MCI), 96 with Alzheimer's disease (AD), and 145 normal controls (NC) scanned as part of the Alzheimer's Disease Neuroimaging Initiative (ADNI). MCI subjects with positive maternal history of dementia had smaller hippocampal volumes at baseline and at follow-up, and greater 12-month atrophy rates than subjects with negative maternal history. Three-dimensional maps and volumetric multiple regression analyses demonstrated a significant effect of positive maternal history of dementia on hippocampal atrophy in MCI and AD after controlling for age, ApoE4 genotype, and paternal history of dementia, respectively. ApoE4 showed an independent effect on hippocampal atrophy in MCI and AD and in the pooled sample.

Cortical and hippocampal atrophy in patients with autosomal dominant familial Alzheimer's disease.

BACKGROUND: Both familial and sporadic Alzheimer's disease (AD) result in progressive cortical and subcortical atrophy. Familial autosomal dominant AD (FAD) allows us to study AD brain changes presymptomatically. METHODS: 33 subjects at risk for FAD (25 for PSEN1 and 8 for APP mutations; 22 mutation carriers and 11 controls) and 3 demented PSEN1 mutation carriers underwent T(1)-weighted MPRAGE 1.5T MRI. Using the hippocampal radial distance and cortical pattern matching techniques, we investigated the effects of carrier status and dementia diagnosis on cortical and hippocampal atrophy. All analyses were corrected for age and relative age (years to median age of disease onset in the family). RESULTS: The dementia cases had pronounced cortical atrophy in the lateral and medial parietal, posterior cingulate and frontal cortices and hippocampal atrophy bilaterally relative to both nondemented carriers and controls. Nondemented carriers did not show significant cortical thinning or hippocampal atrophy relative to controls. CONCLUSIONS: FAD is associated with thinning of the posterior association and frontal cortices and hippocampal atrophy. Larger sample sizes may be necessary to reliably identify cortical atrophy in presymptomatic carriers.

3D PIB and CSF biomarker associations with hippocampal atrophy in ADNI subjects.

Cerebrospinal fluid (CSF) measures of Ab and tau, Pittsburgh Compound B (PIB) imaging and hippocampal atrophy are promising Alzheimer's disease biomarkers yet the associations between them are not known. We applied a validated, automated hippocampal labeling method and 3D radial distance mapping to the 1.5T structural magnetic resonance imaging (MRI) data of 388 ADNI subjects with baseline CSF Ab(42), total tau (t-tau) and phosphorylated tau (p-tau(181)) and 98 subjects with positron emission tomography (PET) imaging using PIB. We used linear regression to investigate associations between hippocampal atrophy and average cortical, parietal and precuneal PIB standardized uptake value ratio (SUVR) and CSF Ab(42), t-tau, p-tau(181), t-tau/Ab(42) and p-tau(181)/Ab(42). All CSF measures showed significant associations with hippocampal volume and radial distance in the pooled sample. Strongest correlations were seen for p-tau(181), followed by p-tau(181)/Ab(42) ratio, t-tau/Ab(42) ratio, t-tau and Ab(42). p-tau(181) showed stronger correlation in ApoE4 carriers, while t-tau showed stronger correlation in ApoE4 noncarriers. Of the 3 PIB measures the precuneal SUVR showed strongest associations with hippocampal atrophy.

Hippocampal, caudate, and ventricular changes in Parkinson's disease with and without dementia.

Parkinson's disease (PD) has been associated with mild cognitive impairment (PDMCI) and with dementia (PDD). Using radial distance mapping, we studied the 3D structural and volumetric differences between the hippocampi, caudates, and lateral ventricles in 20 cognitively normal elderly (NC), 12 cognitively normal PD (PDND), 8 PDMCI, and 15 PDD subjects and examined the associations between these structures and Unified Parkinson's Disease Rating Scale (UPDRS) Part III:motor subscale and Mini-Mental State Examination (MMSE) performance. There were no hippocampal differences between the groups. 3D caudate statistical maps demonstrated significant left medial and lateral and right medial atrophy in the PDD vs. NC, and right medial and lateral caudate atrophy in PDD vs. PDND. PDMCI showed trend-level significant left lateral caudate atrophy vs. NC. Both left and right ventricles were significantly larger in PDD relative to the NC and PDND with posterior (body/occipital horn) predominance. The magnitude of regionally significant between-group differences in radial distance ranged between 20-30% for caudate and 5-20% for ventricles. UPDRS Part III:motor subscale score correlated with ventricular enlargement. MMSE showed significant correlation with expansion of the posterior lateral ventricles and trend-level significant correlation with caudate head atrophy. Cognitive decline in PD is associated with anterior caudate atrophy and ventricular enlargement.

3D comparison of low, intermediate, and advanced hippocampal atrophy in MCI.

We applied the hippocampal radial atrophy mapping technique to the baseline and follow-up magnetic resonance image data of 169 amnestic mild cognitive impairment (MCI) participants in the imaging arm of the Alzheimer's Disease Cooperative Study MCI Donepezil/Vitamin E trial. Sixty percent of the subjects with none to mild hippocampal atrophy rated with the visual medial temporal atrophy rating scale (MTA score < 2) and 33.8% of the subjects with moderate to severe (MTA > or = 2) hippocampal atrophy converted to Alzheimer's disease (AD) during 3-year follow-up. MTA > or = 2 showed a trend for greater left sided hippocampal atrophy versus MTA < 2 groups at baseline (P(corrected) = 0.08). Higher MTA scores were associated with progressive atrophy of the subiculum and the CA1-3 subregions. The MTA < 2 group demonstrated significant bilateral atrophy progression at follow-up (left P(corrected) = 0.008; right P(corrected) = 0.05). Relative to MTA < 2 nonconverters, MTA < 2 converters showed further involvement of the subiculum and CA1 and additional involvement of CA2-3 at follow-up. Right CA1 atrophy was significantly associated with conversion to dementia (for 1 mm greater right CA1 radial distance subjects had 50% reduced hazard for conversion). Greater CA1 and subicular atrophy can be demonstrated early and is predictive of future conversion to AD, whereas CA2-3 involvement becomes more evident as the disease progresses.

Automated 3D mapping of baseline and 12-month associations between three verbal memory measures and hippocampal atrophy in 490 ADNI subjects.

We used a previously validated automated machine learning algorithm based on adaptive boosting to segment the hippocampi in baseline and 12-month follow-up 3D T1-weighted brain MRIs of 150 cognitively normal elderly (NC), 245 mild cognitive impairment (MCI) and 97 Dementia of the Alzheimer's type (DAT) ADNI subjects. Using the radial distance mapping technique, we examined the hippocampal correlates of delayed recall performance on three well-established verbal memory tests--ADAScog delayed recall (ADAScog-DR), the Rey Auditory Verbal Learning Test -DR (AVLT-DR) and Wechsler Logical Memory II-DR (LM II-DR). We observed no significant correlations between delayed recall performance and hippocampal radial distance on any of the three verbal memory measures in NC. All three measures were associated with hippocampal volumes and radial distance in the full sample and in the MCI group at baseline and at follow-up. In DAT we observed stronger left-sided associations between hippocampal radial distance, LM II-DR and ADAScog-DR both at baseline and at follow-up. The strongest linkage between memory performance and hippocampal atrophy in the MCI sample was observed with the most challenging verbal memory test-the AVLT-DR, as opposed to the DAT sample where the least challenging test the ADAScog-DR showed strongest associations with the hippocampal structure. After controlling for baseline hippocampal atrophy, memory performance showed regionally specific associations with hippocampal radial distance in predominantly CA1 but also in subicular distribution.

Subregional hippocampal atrophy predicts Alzheimer's dementia in the cognitively normal.

Atrophic changes of the hippocampus are typically regarded as an early sign of Alzheimer's dementia (AD). Using the radial distance atrophy mapping approach, we compared the longitudinal MRI data of 10 cognitively normal elderly subjects who remained normal at 3-year and 6-year follow-up (NL-NL) and 7 cognitively normal elderly subjects who were diagnosed with mild cognitive impairment (MCI) 2.8 (range 2.0-3.9) and with AD 6.8 years (range 6.1-8.2) after baseline (NL-MCI(AD)). 3D statistical maps revealed greater hippocampal atrophy in the NL-MCI(AD) relative to the NL-NL group at baseline (left p=0.05; right p=0.06) corresponding to 10-15% CA1, and 10-25% subicular atrophy, and bilateral differences at 3-year follow-up (left p=0.001, right p<0.02) corresponding to 10-30% subicular, 10-20% CA1, and 10-20% newly developed CA2-3 atrophy. This preliminary study suggests that excess CA1 and subicular atrophy is present in cognitively normal individuals predestined to decline to amnestic MCI, while progressive involvement of the CA1 and subiculum, and atrophy spreading to the CA2-3 subfield in amnestic MCI, suggests future diagnosis of AD.

Surface feature-guided mapping of cerebral metabolic changes in cognitively normal and mildly impaired elderly.

PURPOSE: The aim of this study was to investigate the longitudinal positron emission tomography (PET) metabolic changes in the elderly. PROCEDURES: Nineteen nondemented subjects (mean Mini-Mental Status Examination 29.4 +/- 0.7 SD) underwent two detailed neuropsychological evaluations and resting 2-deoxy-2-[F-18]fluoro-D: -glucose (FDG)-PET scan (interval 21.7 +/- 3.7 months), baseline structural 3T magnetic resonance (MR) imaging, and apolipoprotein E4 genotyping. Cortical PET metabolic changes were analyzed in 3-D using the cortical pattern matching technique. RESULTS: Baseline vs. follow-up whole-group comparison revealed significant metabolic decline bilaterally in the posterior temporal, parietal, and occipital lobes and the left lateral frontal cortex. The declining group demonstrated 10-15% decline in bilateral posterior cingulate/precuneus, posterior temporal, parietal, and occipital cortices. The cognitively stable group showed 2.5-5% similarly distributed decline. ApoE4-positive individuals underwent 5-15% metabolic decline in the posterior association cortices. CONCLUSIONS: Using 3-D surface-based MR-guided FDG-PET mapping, significant metabolic changes were seen in five posterior and the left lateral frontal regions. The changes were more pronounced for the declining relative to the cognitively stable group.

Comparison of AdaBoost and support vector machines for detecting Alzheimer's disease through automated hippocampal segmentation.

We compared four automated methods for hippocampal segmentation using different machine learning algorithms: 1) hierarchical AdaBoost, 2) support vector machines (SVM) with manual feature selection, 3) hierarchical SVM with automated feature selection (Ada-SVM), and 4) a publicly available brain segmentation package (FreeSurfer). We trained our approaches using T1-weighted brain MRIs from 30 subjects [10 normal elderly, 10 mild cognitive impairment (MCI), and 10 Alzheimer's disease (AD)], and tested on an independent set of 40 subjects (20 normal, 20 AD). Manually segmented gold standard hippocampal tracings were available for all subjects (training and testing). We assessed each approach's accuracy relative to manual segmentations, and its power to map AD effects. We then converted the segmentations into parametric surfaces to map disease effects on anatomy. After surface reconstruction, we computed significance maps, and overall corrected p-values, for the 3-D profile of shape differences between AD and normal subjects. Our AdaBoost and Ada-SVM segmentations compared favorably with the manual segmentations and detected disease effects as well as FreeSurfer on the data tested. Cumulative p-value plots, in conjunction with the false discovery rate method, were used to examine the power of each method to detect correlations with diagnosis and cognitive scores. We also evaluated how segmentation accuracy depended on the size of the training set, providing practical information for future users of this technique.

Automated 3D mapping of hippocampal atrophy and its clinical correlates in 400 subjects with Alzheimer's disease, mild cognitive impairment, and elderly controls.

We used a new method we developed for automated hippocampal segmentation, called the auto context model, to analyze brain MRI scans of 400 subjects from the Alzheimer's disease neuroimaging initiative. After training the classifier on 21 hand-labeled expert segmentations, we created binary maps of the hippocampus for three age- and sex-matched groups: 100 subjects with Alzheimer's disease (AD), 200 with mild cognitive impairment (MCI) and 100 elderly controls (mean age: 75.84; SD: 6.64). Hippocampal traces were converted to parametric surface meshes and a radial atrophy mapping technique was used to compute average surface models and local statistics of atrophy. Surface-based statistical maps visualized links between regional atrophy and diagnosis (MCI versus controls: P = 0.008; MCI versus AD: P = 0.001), mini-mental state exam (MMSE) scores, and global and sum-of-boxes clinical dementia rating scores (CDR; all P < 0.0001, corrected). Right but not left hippocampal atrophy was associated with geriatric depression scores (P = 0.004, corrected); hippocampal atrophy was not associated with subsequent decline in MMSE and CDR scores, educational level, ApoE genotype, systolic or diastolic blood pressure measures, or homocysteine. We gradually reduced sample sizes and used false discovery rate curves to examine the method's power to detect associations with diagnosis and cognition in smaller samples. Forty subjects were sufficient to discriminate AD from normal and correlate atrophy with CDR scores; 104, 200, and 304 subjects, respectively, were required to correlate MMSE with atrophy, to distinguish MCI from normal, and MCI from AD.

Automated mapping of hippocampal atrophy in 1-year repeat MRI data from 490 subjects with Alzheimer's disease, mild cognitive impairment, and elderly controls.

As one of the earliest structures to degenerate in Alzheimer's disease (AD), the hippocampus is the target of many studies of factors that influence rates of brain degeneration in the elderly. In one of the largest brain mapping studies to date, we mapped the 3D profile of hippocampal degeneration over time in 490 subjects scanned twice with brain MRI over a 1-year interval (980 scans). We examined baseline and 1-year follow-up scans of 97 AD subjects (49 males/48 females), 148 healthy control subjects (75 males/73 females), and 245 subjects with mild cognitive impairment (MCI; 160 males/85 females). We used our previously validated automated segmentation method, based on AdaBoost, to create 3D hippocampal surface models in all 980 scans. Hippocampal volume loss rates increased with worsening diagnosis (normal=0.66%/year; MCI=3.12%/year; AD=5.59%/year), and correlated with both baseline and interval changes in Mini-Mental State Examination (MMSE) scores and global and sum-of-boxes Clinical Dementia Rating scale (CDR) scores. Surface-based statistical maps visualized a selective profile of ongoing atrophy in all three diagnostic groups. Healthy controls carrying the ApoE4 gene atrophied faster than non-carriers, while more educated controls atrophied more slowly; converters from MCI to AD showed faster atrophy than non-converters. Hippocampal loss rates can be rapidly mapped, and they track cognitive decline closely enough to be used as surrogate markers of Alzheimer's disease in drug trials. They also reveal genetically greater atrophy in cognitively intact subjects.

Automatic subcortical segmentation using a contextual model.

Automatically segmenting subcortical structures in brain im ages has the potential to greatly accelerate drug trials and population studies of disease. Here we propose an automatic subcortical segmentation algorithm using the auto context model. Unlike many segmentation algorithms that separately compute a shape prior and an image appearance model, we develop a framework based on machine learning to learn a unified appearance and context model. We trained our algorithm to segment the hippocampus and tested it on 83 brain MRIs (of 35 Alzheimer's disease patients, 22 with mild cognitive impairment, and 26 normal healthy controls). Using standard distance and overlap metrics, the auto context model method significantly outperformed simpler learning-based algorithms (using AdaBoost alone) and the FreeSurfer system. In tests on a public domain dataset designed to validate segmentation [1], our new algorithm also greatly improved upon a recently-proposed hybrid discriminative/generative approach [2], which was among the top three that performed comparably in a recent head-to-head competition.

Validation of a fully automated 3D hippocampal segmentation method using subjects with Alzheimer's disease mild cognitive impairment, and elderly controls.

We introduce a new method for brain MRI segmentation, called the auto context model (ACM), to segment the hippocampus automatically in 3D T1-weighted structural brain MRI scans of subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI). In a training phase, our algorithm used 21 hand-labeled segmentations to learn a classification rule for hippocampal versus non-hippocampal regions using a modified AdaBoost method, based on approximately 18,000 features (image intensity, position, image curvatures, image gradients, tissue classification maps of gray/white matter and CSF, and mean, standard deviation, and Haar filters of size 1x1x1 to 7x7x7). We linearly registered all brains to a standard template to devise a basic shape prior to capture the global shape of the hippocampus, defined as the pointwise summation of all the training masks. We also included curvature, gradient, mean, standard deviation, and Haar filters of the shape prior and the tissue classified images as features. During each iteration of ACM - our extension of AdaBoost - the Bayesian posterior distribution of the labeling was fed back in as an input, along with its neighborhood features as new features for AdaBoost to use. In validation studies, we compared our results with hand-labeled segmentations by two experts. Using a leave-one-out approach and standard overlap and distance error metrics, our automated segmentations agreed well with human raters; any differences were comparable to differences between trained human raters. Our error metrics compare favorably with those previously reported for other automated hippocampal segmentations, suggesting the utility of the approach for large-scale studies.