Search

In this study, our objective was to assess the performance of two deep learning‐based hippocampal segmentation methods, SynthSeg and TigerBx, which are readily available to the public. We contrasted their performance with that of two established techniques, FreeSurfer‐Aseg and FSL‐FIRST, using three‐dimensional T1‐weighted MRI scans (n = 1447) procured from public databases. Our evaluation focused on the accuracy and reproducibility of these tools in estimating hippocampal volume. The findings suggest that both SynthSeg and TigerBx are on a par with Aseg and FIRST in terms of segmentation accuracy and reproducibility, but offer a significant advantage in processing speed, generating results in less than 1 min compared with several minutes to hours for the latter tools. In terms of Alzheimer's disease classification based on the hippocampal atrophy rate, SynthSeg and TigerBx exhibited superior performance. In conclusion, we evaluated the capabilities of two deep learning‐based segmentation techniques. The results underscore their potential value in clinical and research environments, particularly when investigating neurological conditions associated with hippocampal structures. [ABSTRACT FROM AUTHOR]

ObjectiveThe pathologic validation of European Alzheimer's Disease Consortium Alzheimer's Disease Neuroimaging Initiative Center Harmonized Hippocampal Segmentation Protocol (HarP).MethodsTemporal lobes of nine Alzheimer's disease (AD) and seven cognitively normal subjects were scanned post-mortem at 7 Tesla. Hippocampal volumes were obtained with HarP. Six-micrometer-thick hippocampal slices were stained for amyloid beta (Aβ), tau, and cresyl violet. Hippocampal subfields were manually traced. Neuronal counts, Aβ, and tau burden for each hippocampal subfield were obtained.ResultsWe found significant correlations between hippocampal volume and Braak and Braak staging (ρ = -0.75, P = .001), tau (ρ = -0.53, P = .034), Aβ burden (ρ = -0.61, P = .012), and neuronal count (ρ = 0.77, P < .001). Exploratory subfield-wise significant associations were found for Aβ in Cornu Ammonis (CA)1 (ρ = -0.58, P = .019) and subiculum (ρ = -0.75, P = .001), tau in CA2 (ρ = -0.59, P = .016), and CA3 (ρ = -0.5, P = .047), and neuronal count in CA1 (ρ = 0.55, P = .028), CA3 (ρ = 0.65, P = .006), and CA4 (ρ = 0.76, P = .001).ConclusionsThe observed associations provide pathological confirmation of hippocampal morphometry as a valid biomarker for AD and pathologic validation of HarP.

Hippocampal segmentation from infant brain MR images is indispensable for studying early brain development. However, most of the hippocampal segmentation methods were developed for population-based adult brain images, which are not suitable for longitudinal infant brain images acquired in the first year of life due to the low image contrast and variable development patterns of the hippocampal structure. To address these challenges, we propose a classification-guided boundary regression method to first detect hippocampal boundaries in the longitudinal infant brain images and then use those detected boundaries to guide the deformable model for final segmentation. Specifically, we first employ a classification-guided regression forest to predict the 3D displacements from individual image voxels to the potential hippocampal boundaries. These predicted displacements then determine the boundary maps by a voting strategy. Second, we iteratively enhance the voted hippocampal boundary map by incorporating the spatial context information given the tentative boundary estimation of the current time point. Besides, the longitudinal context information from all time points of the temporal sequence of the same subject (i.e., given their tentative segmentation results) is also utilized to facilitate accurate segmentation. Finally, a deformable model is applied to the enhanced voted boundary maps for achieving the longitudinal hippocampal segmentation. The experiments on infant brain MR images acquired from 2-week-old to 1-year-old show promising hippocampal segmentation results, indicating the applicability of our method in early brain development study.

Sickle cell disease (SCD) is an inherited hemoglobinopathy that causes organ dysfunction, including cerebral vasculopathy and neurological complications. Hippocampal segmentation with newer and advanced 7 Tesla (7T) MRI protocols has revealed atrophy in specific subregions in other neurodegenerative and neuroinflammatory diseases, however, there is limited evidence of hippocampal involvement in SCD. Thus, we explored whether SCD may be also associated with abnormalities in hippocampal subregions. We conducted 7T MRI imaging in individuals with SCD, including the HbSS, HbSC and HbS/beta thalassemia genotypes (n = 53), and healthy race and age-matched controls (n = 47), using a customized head coil. Both T1- and T2-weighted images were used for automatic segmentation of the hippocampal subfields. Individuals with SCD had, on average, significantly smaller volume of the region including the Dentate Gyrus and Cornu Ammonis (CA) 2 and 3 as compared to the control group. Other hippocampal subregions also showed a trend towards smaller volumes in the SCD group. These findings support and extend previous reports of reduced volume in the temporal lobe in SCD patients. Further studies are necessary to investigate the mechanisms that lead to structural changes in the hippocampus subfields and their relationship with cognitive performance in SCD patients.

The morphological analysis of hippocampus is vital to various neurological studies including brain disorders and brain anatomy. To assist doctors in analyzing the shape and volume of the hippocampus, an accurate and automatic hippocampus segmentation method is highly demanded in the clinical practice. Given that fully convolutional networks (FCNs) have made significant contributions in biomedical image segmentation applications, we propose a notably discriminative feature network based on a hierarchical attention mechanism in hippocampal segmentation. First, considering the problem that the hippocampus is a rather small part in MR images, we design a context-aware high-level feature extraction module (CHFEM) to extract high-level features of scale invariance in the encoder stage. Further, we introduce a hierarchical attention mechanism into our segmentation framework. The mechanism is divided into three parts: a low-level feature spatial attention module (LFSAM) is developed to learn the spatial relationship between different pixels on each channel in the low-level stage of the encoder, a high-level feature channel attention module (HFCAM) is to model the semantic information relationship on different channel images in the high-level stage of the encoder, and a cross-connected attention module (CCAM) is designed in the decoder part to further suppress the noisy boundaries of hippocampus and simultaneously utilize the attentional low-level features from the encoder to better guide the high-level hippocampus edge segmentation in the decoder phase. The proposed approach achieves outstanding performance on the ADNI dataset and the Decathlon dataset compared with other semantic segmentation models and existing hippocampal segmentation approaches. Source code is available at https://github.com/LannyShi/Hippocampal-segmentation. [ABSTRACT FROM AUTHOR]

Abstract Background Hippocampal atrophy is a supportive feature for the diagnosis of probable Alzheimer’s disease (AD). However, even for an expert neuroradiologist, tracing the hippocampus and measuring its volume is a time consuming and extremely challenging task. Accordingly, the development of reliable fully-automated segmentation algorithms is of paramount importance. Materials and methods The present study evaluates (i) the precision and the robustness of the novel Hippocampal Unified Multi-Atlas Network (HUMAN) segmentation algorithm and (ii) its clinical reliability for AD diagnosis. For these purposes, we used a mixed cohort of 456 subjects and their T1 weighted magnetic resonance imaging (MRI) brain scans. The cohort included 145 controls (CTRL), 217 mild cognitive impairment (MCI) subjects and 94 AD patients from Alzheimer’s Disease Neuroimaging Initiative (ADNI). For each subject the baseline, repeat, 12 and 24 month follow-up scans were available. Results HUMAN provides hippocampal volumes with a 3% precision; volume measurements effectively reveal AD, with an area under the curve (AUC) AUC1 = 0.08 ± 0.02. Segmented volumes can also reveal the subtler effects present in MCI subjects, AUC2 = 0.76 ± 0.05. The algorithm is stable and reproducible over time, even for 24 month follow-up scans. Conclusions The experimental results demonstrate HUMAN is a precise segmentation algorithm, besides hippocampal volumes, provided by HUMAN, can effectively support the diagnosis of Alzheimer’s disease and become a useful tool for other neuroimaging applications.

Two variants of semantic dementia are recognized based on the laterality of temporal lobe involvement: a left-predominant variant associated with verbal knowledge impairment and a right-predominant variant associated with behavioural changes and non-verbal knowledge loss. This cross-sectional clinicoradiologic study aimed to assess whole hippocampal, subregion, and/or subfield volume loss in semantic dementia versus controls and across its variants. Thirty-five semantic dementia participants and 15 controls from the Neurodegenerative Research Group at Mayo Clinic who had completed 3.0-T volumetric magnetic resonance imaging and 18 F-fluorodeoxyglucose-positron emission tomography were included. Classification as left-predominant ( n = 25) or right-predominant ( n = 10) variant was based on temporal lobe hypometabolism. Volumes of hippocampal subregions (head, body, and tail) and subfields (parasubiculum, presubiculum, subiculum, cornu ammonis 1, cornu ammonis 3, cornu ammonis 4, dentate gyrus, molecular layer, hippocampal-amygdaloid transition area, and fimbria) were obtained using FreeSurfer 7. Subfield volumes were measured separately from head and body subregions. We fit linear mixed-effects models using log-transformed whole hippocampal/subregion/subfield volumes as dependent variables; age, sex, total intracranial volume, hemisphere and a group-by-hemisphere interaction as fixed effects; and subregion/subfield nested within hemisphere as a random effect. Significant results ( P < 0.05) are hereby reported. At the whole hippocampal level, the dominant (predominantly involved) hemisphere of both variants showed 23-27% smaller volumes than controls. The non-dominant (less involved) hemisphere of the right-predominant variant also showed volume loss versus controls and the left-predominant variant. At the subregional level, both variants showed 17-28% smaller dominant hemisphere head, body, and tail than controls, with the right-predominant variant also showing 8-12% smaller non-dominant hemisphere head than controls and left-predominant variant. At the subfield level, the left-predominant variant showed 12-36% smaller volumes across all dominant hemisphere subfields and 14-15% smaller non-dominant hemisphere parasubiculum, presubiculum (head and body), subiculum (head) and hippocampal-amygdaloid transition area than controls. The right-predominant variant showed 16-49% smaller volumes across all dominant hemisphere subfields and 14-22% smaller parasubiculum, presubiculum, subiculum, cornu ammonis 3, hippocampal-amygdaloid transition area (all from the head) and fimbria of non-dominant hemisphere versus controls. Comparison of dominant hemispheres showed 16-29% smaller volumes of the parasubiculum, presubiculum (head) and fimbria in the right-predominant than left-predominant variant; comparison of non-dominant hemispheres showed 12-15% smaller cornu ammonis 3, cornu ammonis 4, dentate gyrus, hippocampal-amygdaloid transition area (all from the head) and cornu ammonis 1, cornu ammonis 3 and cornu ammonis 4 (all from the body) in the right-predominant variant. All hippocampal subregion/subfield volumes are affected in semantic dementia, although some are more affected in both dominant and non-dominant hemispheres of the right-predominant than the left-predominant variant by the time of presentation. Involvement of hippocampal structures is apparently more subregion dependent than subfield dependent, indicating possible superiority of subregion volumes as disease biomarkers., Competing Interests: A.F.C., S.D.W. and N.T.T.P. have nothing to disclose. J.R.D., H.M.C., R.L.U., M.M.M., H.B., V.J.L., C.G.S., J.L.W. and K.A.J. receive research support from the US NIH. K.A.J. is an Associate Editor for the Annals of Clinical and Translational Neurology., (© The Author(s) 2024. Published by Oxford University Press on behalf of the Guarantors of Brain.)

Hippocampal atrophy is one of the main hallmarks of Alzheimer's disease (AD). However, there is still controversy about whether this sign is a robust finding during the early stages of the disease, such as in mild cognitive impairment (MCI) and subjective cognitive decline (SCD). Considering this background, we proposed a new marker for assessing hippocampal atrophy: the local surface roughness (LSR). We tested this marker in a sample of 307 subjects (normal control (NC) = 70, SCD = 87, MCI = 137, AD = 13). In addition, 97 patients with MCI were followed‐up over a 3‐year period and classified as stable MCI (sMCI) (n = 61) or progressive MCI (pMCI) (n = 36). We did not find significant differences using traditional markers, such as normalized hippocampal volumes (NHV), between the NC and SCD groups or between the sMCI and pMCI groups. However, with LSR we found significant differences between the sMCI and pMCI groups and a better ability to discriminate between NC and SCD. The classification accuracy of the LSR for NC and SCD was 68.2%, while NHV had a 57.2% accuracy. In addition, the classification accuracy of the LSR for sMCI and pMCI was 74.3%, and NHV had a 68.3% accuracy. Cox proportional hazards models adjusted for age, sex, and education were used to estimate the relative hazard of progression from MCI to AD based on hippocampal markers and conversion times. The LSR marker showed better prediction of conversion to AD than NHV. These results suggest the relevance of considering the LSR as a new hippocampal marker for the AD continuum. [ABSTRACT FROM AUTHOR]

Hippocampal atrophy measures from magnetic resonance imaging (MRI) are powerful tools for monitoring Alzheimer's disease (AD) progression. In this paper, we introduce a longitudinal image analysis framework based on robust registration and simultaneous hippocampal segmentation and longitudinal marker classification of brain MRI of an arbitrary number of time points. The framework comprises two innovative parts: a longitudinal segmentation and a longitudinal classification step. The results show that both steps of the longitudinal pipeline improved the reliability and the accuracy of the discrimination between clinical groups. We introduce a novel approach to the joint segmentation of the hippocampus across multiple time points; this approach is based on graph cuts of longitudinal MRI scans with constraints on hippocampal atrophy and supported by atlases. Furthermore, we use linear mixed effect (LME) modeling for differential diagnosis between clinical groups. The classifiers are trained from the average residue between the longitudinal marker of the subjects and the LME model. In our experiments, we analyzed MRI-derived longitudinal hippocampal markers from two publicly available datasets (Alzheimer's Disease Neuroimaging Initiative, ADNI and Minimal Interval Resonance Imaging in Alzheimer's Disease, MIRIAD). In test/retest reliability experiments, the proposed method yielded lower volume errors and significantly higher dice overlaps than the cross-sectional approach (volume errors: 1.55% vs 0.8%; dice overlaps: 0.945 vs 0.975). To diagnose AD, the discrimination ability of our proposal gave an area under the receiver operating characteristic (ROC) curve (AUC) = 0.947 for the control vs AD, AUC = 0.720 for mild cognitive impairment (MCI) vs AD, and AUC = 0.805 for the control vs MCI. [ABSTRACT FROM AUTHOR]

Abstract Recently, a group of major international experts have completed a comprehensive effort to efficiently define a harmonized protocol for manual hippocampal segmentation that is optimized for Alzheimer's research (known as the EADC-ADNI Harmonized Protocol (the HarP)). This study compares the HarP with one of the widely used hippocampal segmentation protocols (Pruessner, 2000), based on a single automatic segmentation method trained separately with libraries made from each manual segmentation protocol. The automatic segmentation conformity with the corresponding manual segmentation and the ability to capture Alzheimer's disease related hippocampal atrophy on large datasets are measured to compare the manual protocols. In addition to the possibility of harmonizing different procedures of hippocampal segmentation, our results show that using the HarP, the automatic segmentation conformity with manual segmentation is also preserved (Dice's κ = 0.88 , κ = 0.87 for Pruessner and HarP respectively (p = 0.726 for common training library)). Furthermore, the results show that the HarP can capture the Alzheimer's disease related hippocampal volume differences in large datasets. The HarP-derived segmentation shows large effect size (Cohen's d = 1.5883) in separating Alzheimer's Disease patients versus normal controls (AD:NC) and medium effect size (Cohen's d = 0.5747) in separating stable versus progressive Mild Cognitively Impaired patients (sMCI:pMCI). Furthermore, the area under the ROC curve for a LDA classifier trained based on age, sex and HarP-derived hippocampal volume is 0.8858 for AD:NC, and for 0.6677 sMCI:pMCI. These results show that the harmonized protocol-derived labels can be widely used in clinic and research, as a sensitive and accurate way of delineating the hippocampus. [ABSTRACT FROM AUTHOR]

Background: Hippocampal atrophy is a supportive feature for the diagnosis of probable Alzheimer's disease (AD). However, even for an expert neuroradiologist, tracing the hippocampus and measuring its volume is a time consuming and extremely challenging task. Accordingly, the development of reliable fully-automated segmentation algorithms is of paramount importance.Materials and Methods: The present study evaluates (i) the precision and the robustness of the novel Hippocampal Unified Multi-Atlas Network (HUMAN) segmentation algorithm and (ii) its clinical reliability for AD diagnosis. For these purposes, we used a mixed cohort of 456 subjects and their T1 weighted magnetic resonance imaging (MRI) brain scans. The cohort included 145 controls (CTRL), 217 mild cognitive impairment (MCI) subjects and 94 AD patients from Alzheimer's Disease Neuroimaging Initiative (ADNI). For each subject the baseline, repeat, 12 and 24 month follow-up scans were available.Results: HUMAN provides hippocampal volumes with a 3% precision; volume measurements effectively reveal AD, with an area under the curve (AUC) AUC1 = 0.08 ± 0.02. Segmented volumes can also reveal the subtler effects present in MCI subjects, AUC2 = 0.76 ± 0.05. The algorithm is stable and reproducible over time, even for 24 month follow-up scans.Conclusions: The experimental results demonstrate HUMAN is a precise segmentation algorithm, besides hippocampal volumes, provided by HUMAN, can effectively support the diagnosis of Alzheimer's disease and become a useful tool for other neuroimaging applications. [ABSTRACT FROM AUTHOR]

The hippocampus is one of the first brain structures affected by Alzheimer's disease (AD). While many automatic methods for hippocampal segmentation exist, few studies have compared them on the same data. In this study, we compare four fully automated hippocampal segmentation methods in terms of their conformity with manual segmentation and their ability to be used as an AD biomarker in clinical settings. We also apply error correction to the four automatic segmentation methods, and complete a comprehensive validation to investigate differences between the methods. The effect size and classification performance is measured for AD versus normal control (NC) groups and for stable mild cognitive impairment (sMCI) versus progressive mild cognitive impairment (pMCI) groups. Our study shows that the nonlinear patch-based segmentation method with error correction is the most accurate automatic segmentation method and yields the most conformity with manual segmentation ( κ = 0.894 ). The largest effect size between AD versus NC and sMCI versus pMCI is produced by FreeSurfer with error correction. We further show that, using only hippocampal volume, age, and sex as features, the area under the receiver operating characteristic curve reaches up to 0.8813 for AD versus NC and 0.6451 for sMCI versus pMCI. However, the automatic segmentation methods are not significantly different in their performance. [ABSTRACT FROM AUTHOR]

We provide and evaluate an open-source software solution for automatically hippocampal segmentation from T1-weighted (T1w) magnetic resonance imaging (MRI). The method is applied for measuring the hippocampal volume, which allows discriminate between patients with Alzheimer's disease (AD) or mild cognitive impairment (MCI) and elderly controls (NC). The method is based on a fast patch-based label fusion method, whose selected patches and their weights are calculated from a combination of similarity measures between patches using intensity-based distances and labeling-based distances. These combined similarity measures produces better selection of the patches, and their weights are more robust. The algorithm is trained with the Harmonized Hippocampal Protocol (HarP). The proposal is compared with FreeSurfer and other label fusion methods. To evaluate the performance and the robustness of the proposed label fusion method, we employ two databases of T1w MRI of human brains. For AD vs NC, we obtain a high degree of accuracy, approximately 90 %. For MCI vs NC, we obtain accuracies around 75 %. The average time for the hippocampal segmentation from a T1w MRI is less than 17 minutes. [ABSTRACT FROM AUTHOR]

Background We provide and evaluate an open-source software solution for automatically measuring hippocampal volume and hippocampal surface roughness based on T1-weighted MRI, which allows for discriminating between patients with Alzheimer's disease (AD) or mild cognitive impairment (MCI) and elderly controls (NC) using only one scan. New method This solution is based on a fast multiple-atlas segmentation technique, which combines a patch-based labeling method with an atlas-warping using non-rigid registrations. Results The classifications are comparable to the best classifications in a large clinical dataset. For AD vs control, we obtain a high degree of accuracy, approximately 90%. For MCI vs control, we obtain accuracies ranging from 70% to 78%. The average time for the hippocampal segmentation from a T1-MRI is less than 17 min. Comparison with existing method In this study, we investigate a combination of our method with annotations using the Harmonized Hippocampal Protocol (HarP). We compare its capabilities with the FreeSurfer method and verify its impact on segmentation and diagnostic group separation capabilities. Our approach is developed and validated using 134 subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database with annotations from HarP. Then, this method, tuned with the best parameters, is applied to 162 subjects from a private image database. Conclusions Our approach with HarP annotations has a high level of accuracy for segmentation of the hippocampus and is robust to multi-site data. The bio-markers extracted from our proposed method have discriminative power based on a scalar feature, showing robustness in generalization and avoid overfitting. The computational time in our hippocampal segmentation algorithm has decreased considerably compared to other available analysis. [ABSTRACT FROM AUTHOR]

Aims Automated hippocampal segmentation is an important issue in many neuroscience studies. Methods We presented and evaluated a novel segmentation method that utilized a manifold learning technique under the multiatlas-based segmentation scenario. A manifold representation of local patches for each voxel was achieved by applying an Isomap algorithm, which can then be used to obtain spatially local weights of atlases for label fusion. The obtained atlas weights potentially depended on all pairwise similarities of the population, which is in contrast to most existing label fusion methods that only rely on similarities between the target image and the atlases. The performance of the proposed method was evaluated for hippocampal segmentation and compared with two representative local weighted label fusion methods, that is, local majority voting and local weighted inverse distance voting, on an in-house dataset of 28 healthy adolescents (age range: 10-17 years) and two ADNI datasets of 100 participants (age range: 60-89 years). We also implemented hippocampal volumetric analysis and evaluated segmentation performance using atlases from a different dataset. Results The median Dice similarities obtained by our proposed method were approximately 0.90 for healthy subjects and above 0.88 for two mixed diagnostic groups of ADNI subjects. Conclusion The experimental results demonstrated that the proposed method could obtain consistent and significant improvements over label fusion strategies that are implemented in the original space. [ABSTRACT FROM AUTHOR]

Objective The objective of this study is to develop a probabilistic modeling framework for segmenting structures of interest from a collection of atlases. We present a label fusion method that is based on minimizing an energy function using graph-cut techniques. Methods and materials We use a conditional random field (CRF) model that allows us to efficiently incorporate shape, appearance and context information. This model is characterized by a pseudo-Boolean function defined on unary, pairwise and higher-order potentials. Given a subset of registered atlases in the target image for a particular region of interest (ROI), we first derive an appearance-shape model from these registered atlases. The unary potentials combine an appearance model based on multiple features with a label prior using a weighted voting method. The pairwise terms are defined from a Finsler metric that minimizes the surface of separation between voxels whose labels are different. The higher-order potentials used in our framework are based on the robust P n model proposed by Kohli et al. The higher-order potentials enforce label consistency in cliques; hence, the proposed method can be viewed as an approach to integrate high-level information with images based on low-level features. To evaluate the performance and the robustness of the proposed label fusion method, we employ two available databases of T1-weighted (T1W) magnetic resonance (MR) images of human brains. We compare our approach with other label fusion methods in the automatic hippocampal segmentation from T1W-MR images. Results Our label fusion method yields mean Dice coefficients of 0.829 and 0.790 for the two databases used with mean times of approximately 80 and 160 s, respectively. Conclusions We introduce a new label fusion method based on a CRF model and on ROIs. The CRF model is characterized by a pseudo-Boolean function defined on unary, pairwise and higher-order potentials. The proposed Boolean function is representable by graphs. A globally optimal binary labeling is found using a st-mincut algorithm in each ROI. We show that the proposed approach is very competitive with respect to recently reported methods. [ABSTRACT FROM AUTHOR]

Background The use of hippocampal volumetry as a biomarker for Alzheimer’s disease (AD) requires that tracers from different laboratories comply with the same segmentation method. Here we present a platform for training and qualifying new tracers to perform the manual segmentation of the hippocampus on magnetic resonance images (MRI) following the European Alzheimer's Disease Consortium and Alzheimer's Disease Neuroimaging Initiative (EADC-ADNI) Harmonized Protocol (HarP). Our objective was to demonstrate that the training process embedded in the platform leads to increased compliance and qualification with the HarP. Method Thirteen new tracers' segmentations were compared with benchmark images with respect to: (a) absolute segmentation volume; (b) spatial overlap of contour with the reference using the Jaccard similarity index; and (c) spatial distance of contour with the reference. Point by point visual feedback was provided through three training phases on 10 MRI. Tracers were then tested on 10 different MRIs in the qualification phase. Results Statistical testing of training over three phases showed a significant increase of Jaccard (i.e. mean Jaccard overlap P < .001) between phases on average for all raters, demonstrating that training positively increased compliance with the HarP. Based on these results we defined qualification thresholds which all tracers were able to meet. Conclusions This platform is an adequate infrastructure allowing standardized training and evaluation of tracers' compliance with the HarP. This is a necessary step allowing the use of hippocampal volumetry as a biomarker for AD in clinical and research centers. [ABSTRACT FROM AUTHOR]

Background The European Alzheimer's Disease Consortium and Alzheimer's Disease Neuroimaging Initiative (ADNI) Harmonized Protocol (HarP) is a Delphi definition of manual hippocampal segmentation from magnetic resonance imaging (MRI) that can be used as the standard of truth to train new tracers, and to validate automated segmentation algorithms. Training requires large and representative data sets of segmented hippocampi. This work aims to produce a set of HarP labels for the proper training and certification of tracers and algorithms. Methods Sixty-eight 1.5 T and 67 3 T volumetric structural ADNI scans from different subjects, balanced by age, medial temporal atrophy, and scanner manufacturer, were segmented by five qualified HarP tracers whose absolute interrater intraclass correlation coefficients were 0.953 and 0.975 (left and right). Labels were validated as HarP compliant through centralized quality check and correction. Results Hippocampal volumes (mm 3 ) were as follows: controls: left = 3060 (standard deviation [SD], 502), right = 3120 (SD, 897); mild cognitive impairment (MCI): left = 2596 (SD, 447), right = 2686 (SD, 473); and Alzheimer’s disease (AD): left = 2301 (SD, 492), right = 2445 (SD, 525). Volumes significantly correlated with atrophy severity at Scheltens' scale (Spearman's ρ = <−0.468, P = <.0005). Cerebrospinal fluid spaces (mm 3 ) were as follows: controls: left = 23 (32), right = 25 (25); MCI: left = 15 (13), right = 22 (16); and AD: left = 11 (13), right = 20 (25). Five subjects (3.7%) presented with unusual anatomy. Conclusions This work provides reference hippocampal labels for the training and certification of automated segmentation algorithms. The publicly released labels will allow the widespread implementation of the standard segmentation protocol. [ABSTRACT FROM AUTHOR]

Automatic and reliable segmentation of subcortical structures is an important but difficult task in quantitative brain image analysis. Multi-atlas based segmentation methods have attracted great interest due to their promising performance. Under the multi-atlas based segmentation framework, using deformation fields generated for registering atlas images onto a target image to be segmented, labels of the atlases are first propagated to the target image space and then fused to get the target image segmentation based on a label fusion strategy. While many label fusion strategies have been developed, most of these methods adopt predefined weighting models that are not necessarily optimal. In this study, we propose a novel local label learning strategy to estimate the target image's segmentation label using statistical machine learning techniques. In particular, we use a L1-regularized support vector machine (SVM) with a k nearest neighbor ( kNN) based training sample selection strategy to learn a classifier for each of the target image voxel from its neighboring voxels in the atlases based on both image intensity and texture features. Our method has produced segmentation results consistently better than state-of-the-art label fusion methods in validation experiments on hippocampal segmentation of over 100 MR images obtained from publicly available and in-house datasets. Volumetric analysis has also demonstrated the capability of our method in detecting hippocampal volume changes due to Alzheimer's disease. Hum Brain Mapp 35:2674-2697, 2014. © 2013 Wiley Periodicals, Inc. [ABSTRACT FROM AUTHOR]

Purpose Hippocampal sclerosis, a common cause of refractory focal epilepsy, requires hippocampal volumetry for accurate diagnosis and surgical planning. Manual segmentation is time-consuming and subject to interrater/intrarater variability. Automated algorithms perform poorly in patients with temporal lobe epilepsy. We validate and make freely available online a novel automated method. Methods Manual hippocampal segmentation was performed on 876, 3 T MRI scans and 202, 1.5 T scans. A template database of 400 high-quality manual segmentations was used to perform automated segmentation of all scans with a multi-atlas-based segmentation propagation method adapted to perform label fusion based on local similarity to ensure accurate segmentation regardless of pathology. Agreement between manual and automated segmentations was assessed by degree of overlap ( Dice coefficient) and comparison of hippocampal volumes. Key Findings The automated segmentation algorithm provided robust delineation of the hippocampi on 3T scans with no more variability than that seen between different human raters ( Dice coefficients: interrater 0.832, manual vs. automated 0.847). In addition, the algorithm provided excellent results with the 1.5T scans ( Dice coefficient 0.827), and automated segmentation remained accurate even in small sclerotic hippocampi. There was a strong correlation between manual and automated hippocampal volumes ( Pearson correlation coefficient 0.929 on the left and 0.941 on the right in 3 T scans). Significance We demonstrate reliable identification of hippocampal atrophy in patients with hippocampal sclerosis, which is crucial for clinical management of epilepsy, particularly if surgical treatment is being contemplated. We provide a free online Web-based service to enable hippocampal volumetry to be available globally, with consequent greatly improved evaluation of those with epilepsy. [ABSTRACT FROM AUTHOR]

Multi-atlas segmentation is an effective approach for automatically labeling objects of interest in biomedical images. In this approach, multiple expert-segmented example images, called atlases, are registered to a target image, and deformed atlas segmentations are combined using label fusion. Among the proposed label fusion strategies, weighted voting with spatially varying weight distributions derived from atlas-target intensity similarity have been particularly successful. However, one limitation of these strategies is that the weights are computed independently for each atlas, without taking into account the fact that different atlases may produce similar label errors. To address this limitation, we propose a new solution for the label fusion problem in which weighted voting is formulated in terms of minimizing the total expectation of labeling error and in which pairwise dependency between atlases is explicitly modeled as the joint probability of two atlases making a segmentation error at a voxel. This probability is approximated using intensity similarity between a pair of atlases and the target image in the neighborhood of each voxel. We validate our method in two medical image segmentation problems: hippocampus segmentation and hippocampus subfield segmentation in magnetic resonance (MR) images. For both problems, we show consistent and significant improvement over label fusion strategies that assign atlas weights independently. [ABSTRACT FROM AUTHOR]

Abstract: We propose a simple but generally applicable approach to improving the accuracy of automatic image segmentation algorithms relative to manual segmentations. The approach is based on the hypothesis that a large fraction of the errors produced by automatic segmentation are systematic, i.e., occur consistently from subject to subject, and serves as a wrapper method around a given host segmentation method. The wrapper method attempts to learn the intensity, spatial and contextual patterns associated with systematic segmentation errors produced by the host method on training data for which manual segmentations are available. The method then attempts to correct such errors in segmentations produced by the host method on new images. One practical use of the proposed wrapper method is to adapt existing segmentation tools, without explicit modification, to imaging data and segmentation protocols that are different from those on which the tools were trained and tuned. An open-source implementation of the proposed wrapper method is provided, and can be applied to a wide range of image segmentation problems. The wrapper method is evaluated with four host brain MRI segmentation methods: hippocampus segmentation using FreeSurfer (); hippocampus segmentation using multi-atlas label fusion (); brain extraction using BET (); and brain tissue segmentation using FAST (). The wrapper method generates 72%, 14%, 29% and 21% fewer erroneously segmented voxels than the respective host segmentation methods. In the hippocampus segmentation experiment with multi-atlas label fusion as the host method, the average Dice overlap between reference segmentations and segmentations produced by the wrapper method is 0.908 for normal controls and 0.893 for patients with mild cognitive impairment. Average Dice overlaps of 0.964, 0.905 and 0.951 are obtained for brain extraction, white matter segmentation and gray matter segmentation, respectively. [Copyright &y& Elsevier]

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 Ti-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. [ABSTRACT FROM AUTHOR]

• The contextual information comes from multi-scale input module and multi-resolution feature fusion module. • Dense connections can further extend the contextual information. • The proposed DDCNet shows significant performance on a limited dataset for hippocampal segmentation. The morphological analysis of hippocampus based on magnetic resonance imaging is significant for computer-aided neurological disease diagnosis. The segmentation of the hippocampus is a prerequisite for calculating its volume and measuring its shape. In this study, we propose a dual dense context-aware network (DDCNet) for automatic hippocampal segmentation, mainly including a multi-scale input approach and multi-resolution feature fusion. First, considering that a single-scale network cannot accurately segment the hippocampus and background with large difference in volume, we design a multi-scale input module (MSIM) to extract the hippocampus features of different receptive fields. Specifically, large scales are used to learn spatial dependencies, while smaller scales are used to learn more hippocampal details. Taking advantage of the fact that the features from different resolution layers offer semantically different information for segmentation, we develop a multi-resolution feature fusion module (MRFFM) to effectively aggregate complementary features and to simultaneously utilize the high-resolution features from the encoder to guide the low-resolution hippocampus edge segmentation in the decoder phase. In addition, many dense connections are introduced to further extend the feature fusion and contextual information of various resolutions between the encoder and decoder. The proposed approach achieves state-of-the-art performance, outperforming existing hippocampal segmentation approaches on the ADNI dataset. [ABSTRACT FROM AUTHOR]