Metric Learning for Multi-atlas based Segmentation of Hippocampus

Deep convolutional neural network for hippocampus segmentation with boundary region refinement

Accurately segmenting the hippocampus from magnetic resonance (MR) brain images is a crucial step in studying brain disorders. However, this task is challenging due to the low signal contrast of hippocampal images, the irregular shape, and small structural size of the hippocampi. In recent years, several deep convolutional networks have been proposed for hippocampus segmentation, which have achieved state-of-the-art performance. These methods typically use large image patches for training the network, as larger patches are beneficial for capturing long-range contextual information. However, this approach increases the computational burden and overlooks the significance of the boundary region. In this study, we propose a deep learning-based method for hippocampus segmentation with boundary region refinement. Our method involves two main steps. First, we propose a convolutional network that takes large image patches as input for initial segmentation. Then, we extract small image patches around the hippocampal boundary for training the second convolutional neural network, which refines the segmentation in the boundary regions. We validate our proposed method on a publicly available dataset and demonstrate that it significantly improves the performance of convolutional neural networks that use single-size image patches as input. In conclusion, our study proposes a novel method for hippocampus segmentation, which improves upon the current state-of-the-art methods. By incorporating a boundary refinement step, our approach achieves higher accuracy in hippocampus segmentation and may facilitate research on brain disorders.

HGM-cNet: Integrating hippocampal gray matter probability map into a cascaded deep learning framework improves hippocampus segmentation

A robust cascaded deep learning framework with integrated hippocampal gray matter (HGM) probability map was developed to improve the hippocampus segmentation (called HGM-cNet) due to its significance in various neuropsychiatric disorders such as Alzheimer's disease (AD). Particularly, the HGM-cNet cascaded two identical convolutional neural networks (CNN), where each CNN was devised by incorporating Attention Block, Residual Block, and DropBlock into the typical encoder-decoder architecture. The two CNNs were skip-connected between encoder components at each scale. The adoption of the cascaded deep learning framework was to conveniently incorporate the HGM probability map with the feature map generated by the first CNN. Experiments on 135T1-weighted MRI scans and manual hippocampal labels from publicly available ADNI-HarP dataset demonstrated that the proposed HGM-cNet outperformed seven multi-atlas-based hippocampus segmentation methods and six deep learning methods under comparison in most evaluation metrics. The Dice (average > 0.89 for both left and right hippocampus) was increased by around or more than 1% over other methods. The HGM-cNet also achieved a superior hippocampus segmentation performance in each group of cognitive normal, mild cognitive impairment, and AD. The stability, conveniences and generalizability of the cascaded deep learning framework with integrated HGM probability map in improving hippocampus segmentation was validated by replacing the proposed CNN with 3D-UNet, Atten-UNet, HippoDeep, QuickNet, DeepHarp, and TransBTS models. The integration of the HGM probability map in the cascaded deep learning framework was also demonstrated to facilitate capturing hippocampal atrophy more accurately than alternative methods in AD analysis. The codes are publicly available at https://github.com/Liu1436510768/HGM-cNet.git.

How segmentation methods affect hippocampal radiomic feature accuracy in Alzheimer's disease analysis?

OBJECTIVES

Hippocampal radiomic features (HRFs) can serve as biomarkers in Alzheimer's disease (AD). However, how different hippocampal segmentation methods affect HRFs in AD is still unknown. The aim of the study was to investigate how different segmentation methods affect HRF accuracy in AD analysis.

METHODS

A total of 1650 subjects were identified from the Alzheimer's Disease Neuroimaging Initiative database (ADNI). The mini-mental state examination (MMSE) and Alzheimer's disease assessment scale (ADAS-cog13) were also adopted. After calculating the HRFs of intensity, shape, and textural features from each side of the hippocampus in structural magnetic resonance imaging (sMRI), the consistency of HRFs calculated from 7 different hippocampal segmentation methods was validated, and the performance of machine learning-based classification of AD vs. normal control (NC) adopting the different HRFs was also examined. Additional 571 subjects from the European DTI Study on Dementia database (EDSD) were to validate the consistency of results.

RESULTS

Between different segmentations, HRFs showed a high measurement consistency (R > 0.7), a high significant consistency between NC, mild cognitive impairment (MCI), and AD (T-value plot, R > 0.8), and consistent significant correlations between HRFs and MMSE/ADAS-cog13 (p < 0.05). The best NC vs. AD classification was obtained when the hippocampus was sufficiently segmented by primitive majority voting (threshold = 0.2). High consistent results were reproduced from independent EDSD cohort.

CONCLUSIONS

HRFs exhibited high consistency across different hippocampal segmentation methods, and the best performance in AD classification was obtained when HRFs were extracted by the naïve majority voting method with a more sufficient segmentation and relatively low hippocampus segmentation accuracy.

KEY POINTS

• The hippocampal radiomic features exhibited high measurement/statistical/clinical consistency across different hippocampal segmentation methods. • The best performance in AD classification was obtained when hippocampal radiomics were extracted by the naïve majority voting method with a more sufficient segmentation and relatively low hippocampus segmentation accuracy.

Hippocampus Segmentation Using U-Net Convolutional Network from Brain Magnetic Resonance Imaging (MRI)

Hippocampus is a part of the limbic system in human brain that plays an important role in forming memories and dealing with intellectual abilities. In most of the neurological disorders related to dementia, such as, Alzheimer's disease, hippocampus is one of the earliest affected regions. Because there are no effective dementia drugs, an ambient assisted living approach may help to prevent or slow the progression of dementia. By segmenting and analyzing the size/shape of hippocampus, it may be possible to classify the early dementia stages. Because of complex structure, traditional image segmentation techniques can't segment hippocampus accurately. Machine learning (ML) is a well known tool in medical image processing that can predict and deliver the outcomes accurately by learning from it's previous results. Convolutional Neural Networks (CNN) is one of the most popular ML algorithms. In this work, a U-Net Convolutional Network based approach is used for hippocampus segmentation from 2D brain images. It is observed that, the original U-Net architecture can segment hippocampus with an average performance rate of 93.6%, which outperforms all other discussed state-of-arts. By using a filter size of [Formula: see text], the original U-Net architecture performs a sequence of convolutional processes. We tweaked the architecture further to extract more relevant features by replacing all [Formula: see text] kernels with three alternative kernels of sizes [Formula: see text], [Formula: see text], and [Formula: see text]. It is observed that, the modified architecture achieved an average performance rate of 96.5%, which outperforms the original U-Net model convincingly.

A robust discriminative multi-atlas label fusion method for hippocampus segmentation from MR image

Accurate and automatic segmentation of the hippocampus plays a vital role in the diagnosis and treatment of nervous system diseases. However, due to the anatomical variability of different subjects, the registered atlas images are not always perfectly aligned with the target image. This makes the segmentation of the hippocampus still face great challenges. In this paper, we propose a robust discriminative label fusion method under the multi-atlas framework. It is a patch embedding label fusion method based on conditional random field (CRF) model that integrates the metric learning and the graph cuts by an integrated formulation. Unlike most current label fusion methods with fixed (non-learning) distance metrics, a novel distance metric learning is presented to enhance discriminative observation and embed it into the unary potential function. In particular, Bayesian inference is utilized to extend a classic distance metric learning, in which large margin constraints are instead of pairwise constraints to obtain a more robust distance metric. And the pairwise homogeneity is fully considered in the spatial prior term based on classification labels and voxel intensity. The resulting integrated formulation is globally minimized by the efficient graph cuts algorithm. Further, sparse patch based method is utilized to polish the obtained segmentation results in label space. The proposed method is evaluated on IABA dataset and ADNI dataset for hippocampus segmentation. The Dice scores achieved by our method are 87.2%, 87.8%, 88.2% and 88.9% on left and right hippocampus on both two datasets, while the best Dice scores obtained by other methods are 86.0%, 86.9%, 86.8% and 88.0% on IABA dataset and ADNI dataset respectively. Experiments show that our approach achieves higher accuracy than state-of-the-art methods. We hope the proposed model can be transferred to combine with other promising distance measurement algorithms.

FCN Based Label Correction for Multi-Atlas Guided Organ Segmentation

Segmentation of medical images using multiple atlases has recently gained immense attention due to their augmented robustness against variabilities across different subjects. These atlas-based methods typically comprise of three steps: atlas selection, image registration, and finally label fusion. Image registration is one of the core steps in this process, accuracy of which directly affects the final labeling performance. However, due to inter-subject anatomical variations, registration errors are inevitable. The aim of this paper is to develop a deep learning-based confidence estimation method to alleviate the potential effects of registration errors. We first propose a fully convolutional network (FCN) with residual connections to learn the relationship between the image patch pair (i.e., patches from the target subject and the atlas) and the related label confidence patch. With the obtained label confidence patch, we can identify the potential errors in the warped atlas labels and correct them. Then, we use two label fusion methods to fuse the corrected atlas labels. The proposed methods are validated on a publicly available dataset for hippocampus segmentation. Experimental results demonstrate that our proposed methods outperform the state-of-the-art segmentation methods.

A Combined Deep-Learning and Lattice Boltzmann Model for Segmentation of the Hippocampus in MRI

Segmentation of the hippocampus (HC) in magnetic resonance imaging (MRI) is an essential step for diagnosis and monitoring of several clinical situations such as Alzheimer's disease (AD), schizophrenia and epilepsy. Automatic segmentation of HC structures is challenging due to their small volume, complex shape, low contrast and discontinuous boundaries. The active contour model (ACM) with a statistical shape prior is robust. However, it is difficult to build a shape prior that is general enough to cover all possible shapes of the HC and that suffers the problems of complicated registration of the shape prior and the target object and of low efficiency. In this paper, we propose a semi-automatic model that combines a deep belief network (DBN) and the lattice Boltzmann (LB) method for the segmentation of HC. The training process of DBN consists of unsupervised bottom-up training and supervised training of a top restricted Boltzmann machine (RBM). Given an input image, the trained DBN is utilized to infer the patient-specific shape prior of the HC. The specific shape prior is not only used to determine the initial contour, but is also introduced into the LB model as part of the external force to refine the segmentation. We used a subset of OASIS-1 as the training set and the preliminary release of EADC-ADNI as the testing set. The segmentation results of our method have good correlation and consistency with the manual segmentation results.

Fast and Precise Hippocampus Segmentation Through Deep Convolutional Neural Network Ensembles and Transfer Learning

Automatic segmentation of the hippocampus from 3D magnetic resonance imaging mostly relied on multi-atlas registration methods. In this work, we exploit recent advances in deep learning to design and implement a fully automatic segmentation method, offering both superior accuracy and fast result. The proposed method is based on deep Convolutional Neural Networks (CNNs) and incorporates distinct segmentation and error correction steps. Segmentation masks are produced by an ensemble of three independent models, operating with orthogonal slices of the input volume, while erroneous labels are subsequently corrected by a combination of Replace and Refine networks. We explore different training approaches and demonstrate how, in CNN-based segmentation, multiple datasets can be effectively combined through transfer learning techniques, allowing for improved segmentation quality. The proposed method was evaluated using two different public datasets and compared favorably to existing methodologies. In the EADC-ADNI HarP dataset, the correspondence between the method's output and the available ground truth manual tracings yielded a mean Dice value of 0.9015, while the required segmentation time for an entire MRI volume was 14.8 seconds. In the MICCAI dataset, the mean Dice value increased to 0.8835 through transfer learning from the larger EADC-ADNI HarP dataset.

Multi-atlas label fusion with random local binary pattern features: Application to hippocampus segmentation

Automatic and reliable segmentation of the hippocampus from magnetic resonance (MR) brain images is extremely important in a variety of neuroimage studies. To improve the hippocampus segmentation performance, a local binary pattern based feature extraction method is developed for machine learning based multi-atlas hippocampus segmentation. Under the framework of multi-atlas image segmentation (MAIS), a set of selected atlases are registered to images to be segmented using a non-linear image registration algorithm. The registered atlases are then used as training data to build linear regression models for segmenting the images based on the image features, referred to as random local binary pattern (RLBP), extracted using a novel image feature extraction method. The RLBP based MAIS algorithm has been validated for segmenting hippocampus based on a data set of 135 T1 MR images which are from the Alzheimer’s Disease Neuroimaging Initiative database (adni.loni.usc.edu). By using manual segmentation labels produced by experienced tracers as the standard of truth, six segmentation evaluation metrics were used to evaluate the image segmentation results by comparing automatic segmentation results with the manual segmentation labels. We further computed Cohen’s d effect size to investigate the sensitivity of each segmenting method in detecting volumetric differences of the hippocampus between different groups of subjects. The evaluation results showed that our method was competitive to state-of-the-art label fusion methods in terms of accuracy. Hippocampal volumetric analysis showed that the proposed RLBP method performed well in detecting the volumetric differences of the hippocampus between groups of Alzheimer’s disease patients, mild cognitive impairment subjects, and normal controls. These results have demonstrated that the RLBP based multi-atlas image segmentation method could facilitate efficient and accurate extraction of the hippocampus and may help predict Alzheimer’s disease. The codes of the proposed method is available (https://www.nitrc.org/frs/?group_id=1242).

Hippocampus Segmentation Based on Iterative Local Linear Mapping With Representative and Local Structure-Preserved Feature Embedding

Hippocampus segmentation plays a significant role in mental disease diagnoses, such as Alzheimer's disease, epilepsy, and so on. Patch-based multi-atlas segmentation (PBMAS) approach is a popular method for hippocampus segmentation and has achieved a promising result. However, the PBMAS approach needs high computation cost due to registration and the segmentation accuracy is subject to the registration accuracy. In this paper, we propose a novel method based on iterative local linear mapping (ILLM) with the representative and local structure-preserved feature embedding to achieve accurate and robust hippocampus segmentation with no need for registration. In the proposed approach, semi-supervised deep autoencoder (SSDA) exploits unsupervised deep autoencoder and local structure-preserved manifold regularization to nonlinearly transform the extracted magnetic resonance (MR) patch to embedded feature manifold, whose adjacent relationship is similar to the signed distance map (SDM) patch manifold. Local linear mapping is used to preliminarily predict SDM patch corresponding to the MR patch. Subsequently, threshold segmentation generates a preliminary segmentation. The ILLM refines the segmentation result iteratively by ensuring the local constraints of embedded feature manifold and SDM patch manifold using a space-constrained dictionary update. Thus, a refined segmentation is obtained with no need for registration. The experiments on 135 subjects from ADNI dataset show that the proposed approach is superior to the state-of-the-art PBMAS and classification-based approaches with mean Dice similarity coefficients of 0.8852±0.0203 and 0.8783 ± 0.0251 for bilateral hippocampus segmentation of 1.5T and 3.0T datasets, respectively.

Non-linear registration improves statistical power to detect hippocampal atrophy in aging and dementia

OBJECTIVE

To compare the performance of different methods for determining hippocampal atrophy rates using longitudinal MRI scans in aging and Alzheimer's disease (AD).

BACKGROUND

Quantifying hippocampal atrophy caused by neurodegenerative diseases is important to follow the course of the disease. In dementia, the efficacy of new therapies can be partially assessed by measuring their effect on hippocampal atrophy. In radiotherapy, the quantification of radiation-induced hippocampal volume loss is of interest to quantify radiation damage. We evaluated plausibility, reproducibility and sensitivity of eight commonly used methods to determine hippocampal atrophy rates using test-retest scans.

MATERIALS AND METHODS

Manual, FSL-FIRST, FreeSurfer, multi-atlas segmentation (MALF) and non-linear registration methods (Elastix, NiftyReg, ANTs and MIRTK) were used to determine hippocampal atrophy rates on longitudinal T1-weighted MRI from the ADNI database. Appropriate parameters for the non-linear registration methods were determined using a small training dataset (N = 16) in which two-year hippocampal atrophy was measured using test-retest scans of 8 subjects with low and 8 subjects with high atrophy rates. On a larger dataset of 20 controls, 40 mild cognitive impairment (MCI) and 20  AD patients, one-year hippocampal atrophy rates were measured. A repeated measures ANOVA analysis was performed to determine differences between controls, MCI and AD patients. For each method we calculated effect sizes and the required sample sizes to detect one-year volume change between controls and MCI (NCTRL_MCI) and between controls and AD (NCTRL_AD). Finally, reproducibility of hippocampal atrophy rates was assessed using within-session rescans and expressed as an average distance measure DAve, which expresses the difference in atrophy rate, averaged over all subjects. The same DAve was used to determine the agreement between different methods.

RESULTS

Except for MALF, all methods detected a significant group difference between CTRL and AD, but none could find a significant difference between the CTRL and MCI. FreeSurfer and MIRTK required the lowest sample sizes (FreeSurfer: NCTRL_MCI = 115, NCTRL_AD = 17 with DAve = 3.26%; MIRTK: NCTRL_MCI = 97, NCTRL_AD = 11 with DAve = 3.76%), while ANTs was most reproducible (NCTRL_MCI = 162, NCTRL_AD = 37 with DAve = 1.06%), followed by Elastix (NCTRL_MCI = 226, NCTRL_AD = 15 with DAve = 1.78%) and NiftyReg (NCTRL_MCI = 193, NCTRL_AD = 14 with DAve = 2.11%). Manually measured hippocampal atrophy rates required largest sample sizes to detect volume change and were poorly reproduced (NCTRL_MCI = 452, NCTRL_AD = 87 with DAve = 12.39%). Atrophy rates of non-linear registration methods also agreed best with each other.

DISCUSSION AND CONCLUSION

Non-linear registration methods were most consistent in determining hippocampal atrophy and because of their better reproducibility, methods, such as ANTs, Elastix and NiftyReg, are preferred for determining hippocampal atrophy rates on longitudinal MRI. Since performances of non-linear registration methods are well comparable, the preferred method would mostly depend on computational efficiency.

Adaptive Bayesian label fusion using kernel-based similarity metrics in hippocampus segmentation

The effectiveness of brain magnetic resonance imaging (MRI) as a useful evaluation tool strongly depends on the performed segmentation of associated tissues or anatomical structures. We introduce an enhanced brain segmentation approach of Bayesian label fusion that includes the construction of adaptive target-specific probabilistic priors using atlases ranked by kernel-based similarity metrics to deal with the anatomical variability of collected MRI data. In particular, the developed segmentation approach appraises patch-based voxel representation to enhance the voxel embedding in spaces with increased tissue discrimination, as well as the construction of a neighborhood-dependent model that addresses the label assignment of each region with a different patch complexity. To measure the similarity between the target and training atlases, we propose a tensor-based kernel metric that also includes the training labeling set. We evaluate the proposed approach, adaptive Bayesian label fusion using kernel-based similarity metrics, in the specific case of hippocampus segmentation of five benchmark MRI collections, including ADNI dataset, resulting in an increased performance (assessed through the Dice index) as compared to other recent works.

Integrating Semi-supervised and Supervised Learning Methods for Label Fusion in Multi-Atlas Based Image Segmentation

A novel label fusion method for multi-atlas based image segmentation method is developed by integrating semi-supervised and supervised machine learning techniques. Particularly, our method is developed in a pattern recognition based multi-atlas label fusion framework. We build random forests classification models for each image voxel to be segmented based on its corresponding image patches of atlas images that have been registered to the image to be segmented. The voxelwise random forests classification models are then applied to the image to be segmented to obtain a probabilistic segmentation map. Finally, a semi-supervised label propagation method is adapted to refine the probabilistic segmentation map by propagating its reliable voxelwise segmentation labels, taking into consideration consistency of local and global image appearance of the image to be segmented. The proposed method has been evaluated for segmenting the hippocampus in MR images and compared with alternative machine learning based multi-atlas based image segmentation methods. The experiment results have demonstrated that our method could obtain competitive segmentation performance (average Dice index > 0.88), compared with alternative multi-atlas based image segmentation methods under comparison. Source codes of the methods under comparison are publicly available at www.nitrc.org/frs/?group_id=1242.

INTEGRATING SEMI-SUPERVISED LABEL PROPAGATION AND RANDOM FORESTS FOR MULTI-ATLAS BASED HIPPOCAMPUS SEGMENTATION

A novel multi-atlas based image segmentation method is proposed by integrating a semi-supervised label propagation method and a supervised random forests method in a pattern recognition based label fusion framework. The semi-supervised label propagation method takes into consideration local and global image appearance of images to be segmented and segments the images by propagating reliable segmentation results obtained by the supervised random forests method. Particularly, the random forests method is used to train a regression model based on image patches of atlas images for each voxel of the images to be segmented. The regression model is used to obtain reliable segmentation results to guide the label propagation for the segmentation. The proposed method has been compared with state-of-the-art multi-atlas based image segmentation methods for segmenting the hippocampus in MR images. The experiment results have demonstrated that our method obtained superior segmentation performance.

Inter-observer variation of hippocampus delineation in hippocampal avoidance prophylactic cranial irradiation

BACKGROUND

Hippocampal avoidance prophylactic cranial irradiation (HA-PCI) techniques have been developed to reduce radiation damage to the hippocampus. An inter-observer hippocampus delineation analysis was performed and the influence of the delineation variability on dose to the hippocampus was studied.

MATERIALS AND METHODS

For five patients, seven observers delineated both hippocampi on brain MRI. The intra-class correlation (ICC) with absolute agreement and the generalized conformity index (CI_gen) were computed. Median surfaces over all observers' delineations were created for each patient and regional outlining differences were analysed. HA-PCI dose plans were made from the median surfaces and we investigated whether dose constraints in the hippocampus could be met for all delineations.

RESULTS

The ICC for the left and right hippocampus was 0.56 and 0.69, respectively, while the CI_gen ranged from 0.55 to 0.70. The posterior and anterior-medial hippocampal regions had most variation with SDs ranging from approximately 1 to 2.5 mm. The mean dose (D_mean) constraint was met for all delineations, but for the dose received by 1% of the hippocampal volume (D_1%) violations were observed.

CONCLUSION

The relatively low ICC and CI_gen indicate that delineation variability among observers for both left and right hippocampus was large. The posterior and anterior-medial border have the largest delineation inaccuracy. The hippocampus D_mean constraint was not violated.

Patch-Based Label Fusion with Structured Discriminant Embedding for Hippocampus Segmentation

Automatic and accurate segmentation of hippocampal structures in medical images is of great importance in neuroscience studies. In multi-atlas based segmentation methods, to alleviate the misalignment when registering atlases to the target image, patch-based methods have been widely studied to improve the performance of label fusion. However, weights assigned to the fused labels are usually computed based on predefined features (e.g. image intensities), thus being not necessarily optimal. Due to the lack of discriminating features, the original feature space defined by image intensities may limit the description accuracy. To solve this problem, we propose a patch-based label fusion with structured discriminant embedding method to automatically segment the hippocampal structure from the target image in a voxel-wise manner. Specifically, multi-scale intensity features and texture features are first extracted from the image patch for feature representation. Margin fisher analysis (MFA) is then applied to the neighboring samples in the atlases for the target voxel, in order to learn a subspace in which the distance between intra-class samples is minimized and the distance between inter-class samples is simultaneously maximized. Finally, the k-nearest neighbor (kNN) classifier is employed in the learned subspace to determine the final label for the target voxel. In the experiments, we evaluate our proposed method by conducting hippocampus segmentation using the ADNI dataset. Both the qualitative and quantitative results show that our method outperforms the conventional multi-atlas based segmentation methods.

Patch-based local learning method for cerebral blood flow quantification with arterial spin-labeling MRI

Arterial spin-labeling (ASL) perfusion MRI is a non-invasive method for quantifying cerebral blood flow (CBF). Standard ASL CBF calibration mainly relies on pair-wise subtraction of the spin-labeled images and controls images at each voxel separately, ignoring the abundant spatial correlations in ASL data. To address this issue, we previously proposed a multivariate support vector machine (SVM) learning-based algorithm for ASL CBF quantification (SVMASLQ). But the original SVMASLQ was designed to do CBF quantification for all image voxels simultaneously, which is not ideal for considering local signal and noise variations. To fix this problem, we here in this paper extended SVMASLQ into a patch-wise method by using a patch-wise classification kernel. At each voxel, an image patch centered at that voxel was extracted from both the control images and labeled images, which was then input into SVMASLQ to find the corresponding patch of the surrogate perfusion map using a non-linear SVM classifier. Those patches were eventually combined into the final perfusion map. Method evaluations were performed using ASL data from 30 young healthy subjects. The results showed that the patch-wise SVMASLQ increased perfusion map SNR by 6.6% compared to the non-patch-wise SVMASLQ.