An adaptive mean-shift framework for MRI brain segmentation

Novel Solution for Using Neural Networks for Kidney Boundary Extraction in 2D Ultrasound Data

Background and Objective: Kidney ultrasound (US) imaging is a significant imaging modality for evaluating kidney health and is essential for diagnosis, treatment, surgical intervention planning, and follow-up assessments. Kidney US image segmentation consists of extracting useful objects or regions from the total image, which helps determine tissue organization and improve diagnosis. Thus, obtaining accurate kidney segmentation data is an important first step for precisely diagnosing kidney diseases. However, manual delineation of the kidney in US images is complex and tedious in clinical practice. To overcome these challenges, we developed a novel automatic method for US kidney segmentation. Methods: Our method comprises two cascaded steps for US kidney segmentation. The first step utilizes a coarse segmentation procedure based on a deep fusion learning network to roughly segment each input US kidney image. The second step utilizes a refinement procedure to fine-tune the result of the first step by combining an automatic searching polygon tracking method with a machine learning network. In the machine learning network, a suitable and explainable mathematical formula for kidney contours is denoted by basic parameters. Results: Our method is assessed using 1380 trans-abdominal US kidney images obtained from 115 patients. Based on comprehensive comparisons of different noise levels, our method achieves accurate and robust results for kidney segmentation. We use ablation experiments to assess the significance of each component of the method. Compared with state-of-the-art methods, the evaluation metrics of our method are significantly higher. The Dice similarity coefficient (DSC) of our method is 94.6 ± 3.4%, which is higher than those of recent deep learning and hybrid algorithms (89.4 ± 7.1% and 93.7 ± 3.8%, respectively). Conclusions: We develop a coarse-to-refined architecture for the accurate segmentation of US kidney images. It is important to precisely extract kidney contour features because segmentation errors can cause under-dosing of the target or over-dosing of neighboring normal tissues during US-guided brachytherapy. Hence, our method can be used to increase the rigor of kidney US segmentation.

Automated Detection of Patients at High Risk of Polypharmacy including Anticholinergic and Sedative Medications

Ensuring that medicines are prescribed safely is fundamental to the role of healthcare professionals who need to be vigilant about the risks associated with drugs and their interactions with other medicines (polypharmacy). One aspect of preventative healthcare is to use artificial intelligence to identify patients at risk using big data analytics. This will improve patient outcomes by enabling pre-emptive changes to medication on the identified cohort before symptoms present. This paper presents a mean-shift clustering technique used to identify groups of patients at the highest risk of polypharmacy. A weighted anticholinergic risk score and a weighted drug interaction risk score were calculated for each of 300,000 patient records registered with a major regional UK-based healthcare provider. The two measures were input into the mean-shift clustering algorithm and this grouped patients into clusters reflecting different levels of polypharmaceutical risk. Firstly, the results showed that, for most of the data, the average scores are not correlated and, secondly, the high risk outliers have high scores for one measure but not for both. These suggest that any systematic recognition of high-risk groups should consider both anticholinergic and drug-drug interaction risks to avoid missing high-risk patients. The technique was implemented in a healthcare management system and easily and automatically identifies groups at risk far faster than the manual inspection of patient records. This is much less labour-intensive for healthcare professionals who can focus their assessment only on patients within the high-risk group(s), enabling more timely clinical interventions where necessary.

Synthetic augmentation for semantic segmentation of class imbalanced biomedical images: A data pair generative adversarial network approach

In recent years, deep learning (DL) has been recognized very useful in the semantic segmentation of biomedical images. Such an application, however, is significantly hindered by the lack of pixel-wise annotations. In this work, we propose a data pair generative adversarial network (DPGAN) for the purpose of synthesizing concurrently the diverse biomedical images and the segmentation labels from random latent vectors. First, a hierarchical structure is constructed consisting of three variational auto-encoder generative adversarial networks (VAEGANs) with an extra discriminator. Subsequently, to alleviate the influence from the imbalance between lesions and non-lesions areas in biomedical segmentation data sets, we divide the DPGAN into three stages, namely, background stage, mask stage and advanced stage, with each stage deploying a VAEGAN. In such a way, a large number of new segmentation data pairs are generated from random latent vectors and then used to augment the original data sets. Finally, to validate the effectiveness of the proposed DPGAN, experiments are carried out on a vestibular schwannoma data set, a kidney tumor data set and a skin cancer data set. The results indicate that, in comparison to other state-of-the-art GAN-based methods, the proposed DPGAN shows better performance in the generative quality, and meanwhile, gains an effective boost on semantic segmentation of class imbalanced biomedical images.

Pseudo-Label-Assisted Self-Organizing Maps for Brain Tissue Segmentation in Magnetic Resonance Imaging

Brain tissue segmentation in magnetic resonance imaging volumes is an important image processing step for analyzing the human brain. This paper presents a novel approach named Pseudo-Label Assisted Self-Organizing Map (PLA-SOM) that enhances the result produced by a base segmentation method. Using the output of a base method, PLA-SOM calculates pseudo-labels in order to keep inter-class separation and intra-class compactness in the training phase. For the mapping phase, PLA-SOM uses a novel fuzzy function that combines feature space learned by the SOM's prototypes, topological ordering from the map, and spatial information from a brain atlas. We assessed PLA-SOM performance on synthetic and real MRIs of the brain, obtained from the BrainWeb and the Internet Brain Image Repository datasets. The experimental results showed evidence of segmentation improvement achieved by the proposed method over six different base methods. The best segmentation improvements reported by PLA-SOM on synthetic brain scans are 11%, 6%, and 4% for the tissue classes cerebrospinal fluid, gray matter, and white matter, respectively. On real brain scans, PLA-SOM achieved segmentation enhancements of 15%, 5%, and 12% for cerebrospinal fluid, gray matter, and white matter, respectively.

An atlas of classifiers-a machine learning paradigm for brain MRI segmentation

We present the Atlas of Classifiers (AoC)-a conceptually novel framework for brain MRI segmentation. The AoC is a spatial map of voxel-wise multinomial logistic regression (LR) functions learned from the labeled data. Upon convergence, the resulting fixed LR weights, a few for each voxel, represent the training dataset. It can, therefore, be considered as a light-weight learning machine, which despite its low capacity does not underfit the problem. The AoC construction is independent of the actual intensities of the test images, providing the flexibility to train it on the available labeled data and use it for the segmentation of images from different datasets and modalities. In this sense, it does not overfit the training data, as well. The proposed method has been applied to numerous publicly available datasets for the segmentation of brain MRI tissues and is shown to be robust to noise and outreach commonly used methods. Promising results were also obtained for multi-modal, cross-modality MRI segmentation. Finally, we show how AoC trained on brain MRIs of healthy subjects can be exploited for lesion segmentation of multiple sclerosis patients.

Segmentation of MRI brain scans using spatial constraints and 3D features

This paper presents a novel unsupervised algorithm for brain tissue segmentation in magnetic resonance imaging (MRI). The proposed algorithm, named Gardens2, adopts a clustering approach to segment voxels of a given MRI into three classes: cerebrospinal fluid (CSF), gray matter (GM), and white matter (WM). Using an overlapping criterion, 3D feature descriptors and prior atlas information, Gardens2 generates a segmentation mask per class in order to parcellate the brain tissues. We assessed our method using three neuroimaging datasets: BrainWeb, IBSR18, and IBSR20, the last two provided by the Internet Brain Segmentation Repository. Its performance was compared with eleven well established as well as newly proposed unsupervised segmentation methods. Overall, Gardens2 obtained better segmentation performance than the rest of the methods in two of the three databases and competitive results when its performance was measured by class. Graphical Abstract Brain tissue segmentation using 3D features and an adjusted atlas template.

Alzheimer's diagnosis using deep learning in segmenting and classifying 3D brain MR images

BACKGROUND AND OBJECTIVES

Dementia is one of the brain diseases with serious symptoms such as memory loss, and thinking problems. According to the World Alzheimer Report 2016, in the world, there are 47 million people having dementia and it can be 131 million by 2050. There is no standard method to diagnose dementia, and consequently unable to access the treatment effectively. Hence, the computational diagnosis of the disease from brain Magnetic Resonance Image (MRI) scans plays an important role in supporting the early diagnosis. Alzheimer's Disease (AD), a common type of Dementia, includes problems related to disorientation, mood swings, not managing self-care, and behavioral issues. In this article, we present a new computational method to diagnosis Alzheimer's disease from 3D brain MR images.

METHODS

An efficient approach to diagnosis Alzheimer's disease from brain MRI scans is proposed comprising two phases: I) segmentation and II) classification, both based on deep learning. After the brain tissues are segmented by a model that combines Gaussian Mixture Model (GMM) and Convolutional Neural Network (CNN), a new model combining Extreme Gradient Boosting (XGBoost) and Support Vector Machine (SVM) is used to classify Alzheimer's disease based on the segmented tissues.

RESULTS

We present two evaluations for segmentation and classification. For comparison, the new method was evaluated using the AD-86 and AD-126 datasets leading to Dice 0.96 for segmentation in both datasets and accuracies 0.88, and 0.80 for classification, respectively.

CONCLUSION

Deep learning gives prominent results for segmentation and feature extraction in medical image processing. The combination of XGboost and SVM improves the results obtained.

A Survey on Computer-Aided Diagnosis of Brain Disorders through MRI Based on Machine Learning and Data Mining Methodologies with an Emphasis on Alzheimer Disease Diagnosis and the Contribution of the Multimodal Fusion

Computer-aided diagnostic (CAD) systems use machine learning methods that provide a synergistic effect between the neuroradiologist and the computer, enabling an efficient and rapid diagnosis of the patient’s condition. As part of the early diagnosis of Alzheimer’s disease (AD), which is a major public health problem, the CAD system provides a neuropsychological assessment that helps mitigate its effects. The use of data fusion techniques by CAD systems has proven to be useful, they allow for the merging of information relating to the brain and its tissues from MRI, with that of other types of modalities. This multimodal fusion refines the quality of brain images by reducing redundancy and randomness, which contributes to improving the clinical reliability of the diagnosis compared to the use of a single modality. The purpose of this article is first to determine the main steps of the CAD system for brain magnetic resonance imaging (MRI). Then to bring together some research work related to the diagnosis of brain disorders, emphasizing AD. Thus the most used methods in the stages of classification and brain regions segmentation are described, highlighting their advantages and disadvantages. Secondly, on the basis of the raised problem, we propose a solution within the framework of multimodal fusion. In this context, based on quantitative measurement parameters, a performance study of multimodal CAD systems is proposed by comparing their effectiveness with those exploiting a single MRI modality. In this case, advances in information fusion techniques in medical imagery are accentuated, highlighting their advantages and disadvantages. The contribution of multimodal fusion and the interest of hybrid models are finally addressed, as well as the main scientific assertions made, in the field of brain disease diagnosis.

Robust spatial fuzzy GMM based MRI segmentation and carotid artery plaque detection in ultrasound images

BACKGROUND AND OBJECTIVE

In medical image analysis for disease diagnosis, segmentation is one of the challenging tasks. Owing to the inherited degradations in MRI improper segments are produced. Segmentation process is an important step in brain tissue analysis. Moreover, an early detection of plaque in carotid artery using ultrasound images may prevent serious brain strokes. Unfortunately, low quality and noisy ultrasound images are still challenges for accurate segmentation. The objective of this research is to develop a robust segmentation approach for medical images such as brain MRI and carotid artery ultrasound images.

METHODS

In this paper, a novel approach is proposed to address the segmentation challenges of medical images. The proposed approach employed fuzzy intelligence and Gaussian mixture model (GMM). It comprises two phases; firstly, incorporating spatial fuzzy c-means in GMM by exploiting statistical, texture, and wavelet image features. During model development, GMM parameters are estimated in presence of noise by EM algorithm iteratively. Utilizing these parameters, brain MRI images are segmented. In next phase, developed approach is applied to solve a real problem of carotid artery plaque detection using ultrasound images. The dataset of real patients annotated by radiologists has been obtained from Radiology Department, Shifa International Hospital Islamabad, Pakistan. For this, intima-media-thickness values are computed from the proposed segmentation followed by support vector machines for plaque classification (normal/abnormal).

RESULTS

The obtained segmentation has been evaluated on standard brain MRI dataset and offers high segmentation accuracy of 99.2%. The proposed approach outperforms in term of segmentation performance range of 3-9% as compared to the state of the art approaches on brain MRI. Furthermore, the proposed approach shows robustness to various levels of Gaussian and Rician image noises. On carotid artery dataset, we have obtained high plaque detection rate in terms of accuracy, sensitivity, specificity, and F-score values of 98.8%, 99.3%, 98.0%, and 97.5% respectively.

CONCLUSIONS

The proposed approach segments both modalities with high precision and shows robustness at Gaussian and Rician noise levels. Results for brain MRI and ultrasound images indicate its effectiveness and can be used as second opinion in addition to the radiologists. The developed approach is straightforward, efficient, and reproducible. It may benefit to improve the clinical evaluation of the disease in both asymptomatic and symptomatic individuals.

DCS-SVM: a novel semi-automated method for human brain MR image segmentation

In this paper, a novel method is proposed which appropriately segments magnetic resonance (MR) brain images into three main tissues. This paper proposes an extension of our previous work in which we suggested a combination of multiple classifiers (CMC)-based methods named dynamic classifier selection-dynamic local training local Tanimoto index (DCS-DLTLTI) for MR brain image segmentation into three main cerebral tissues. This idea is used here and a novel method is developed that tries to use more complex and accurate classifiers like support vector machine (SVM) in the ensemble. This work is challenging because the CMC-based methods are time consuming, especially on huge datasets like three-dimensional (3D) brain MR images. Moreover, SVM is a powerful method that is used for modeling datasets with complex feature space, but it also has huge computational cost for big datasets, especially those with strong interclass variability problems and with more than two classes such as 3D brain images; therefore, we cannot use SVM in DCS-DLTLTI. Therefore, we propose a novel approach named "DCS-SVM" to use SVM in DCS-DLTLTI to improve the accuracy of segmentation results. The proposed method is applied on well-known datasets of the Internet Brain Segmentation Repository (IBSR) and promising results are obtained.

Intuitionistic Center-Free FCM Clustering for MR Brain Image Segmentation

In this paper, an intuitionistic center-free fuzzy c-means clustering method (ICFFCM) is proposed for magnetic resonance (MR) brain image segmentation. First, in order to suppress the effect of noise in MR brain images, a pixel-to-pixel similarity with spatial information is defined. Then, for the purpose of handling the vagueness in MR brain images as well as the uncertainty in clustering process, a pixel-to-cluster similarity measure is defined by employing the intuitionistic fuzzy membership function. These two similarities are used to modify the center-free FCM so that the ability of the method for MR brain image segmentation could be improved. Second, on the basis of the improved center-free FCM method, a local information term, which is also intuitionistic and center-free, is appended to the objective function. This generates the final proposed ICFFCM. The consideration of local information further enhances the robustness of ICFFCM to the noise in MR brain images. Experimental results on the simulated and real MR brain image datasets show that ICFFCM is effective and robust. Moreover, ICFFCM could outperform several fuzzy-clustering-based methods and could achieve comparable results to the standard published methods like statistical parametric mapping and FMRIB automated segmentation tool.

Learning Based Segmentation of CT Brain Images: Application to Postoperative Hydrocephalic Scans

OBJECTIVE

Hydrocephalus is a medical condition in which there is an abnormal accumulation of cerebrospinal fluid (CSF) in the brain. Segmentation of brain imagery into brain tissue and CSF [before and after surgery, i.e., preoperative (pre-op) versus postoperative (post-op)] plays a crucial role in evaluating surgical treatment. Segmentation of pre-op images is often a relatively straightforward problem and has been well researched. However, segmenting post-op computational tomographic (CT) scans becomes more challenging due to distorted anatomy and subdural hematoma collections pressing on the brain. Most intensity- and feature-based segmentation methods fail to separate subdurals from brain and CSF as subdural geometry varies greatly across different patients and their intensity varies with time. We combat this problem by a learning approach that treats segmentation as supervised classification at the pixel level, i.e., a training set of CT scans with labeled pixel identities is employed.

METHODS

Our contributions include: 1) a dictionary learning framework that learns class (segment) specific dictionaries that can efficiently represent test samples from the same class while poorly represent corresponding samples from other classes; 2) quantification of associated computation and memory footprint; and 3) a customized training and test procedure for segmenting post-op hydrocephalic CT images.

RESULTS

Experiments performed on infant CT brain images acquired from the CURE Children's Hospital of Uganda reveal the success of our method against the state-of-the-art alternatives. We also demonstrate that the proposed algorithm is computationally less burdensome and exhibits a graceful degradation against a number of training samples, enhancing its deployment potential.

A fast stochastic framework for automatic MR brain images segmentation

This paper introduces a new framework for the segmentation of different brain structures (white matter, gray matter, and cerebrospinal fluid) from 3D MR brain images at different life stages. The proposed segmentation framework is based on a shape prior built using a subset of co-aligned training images that is adapted during the segmentation process based on first- and second-order visual appearance characteristics of MR images. These characteristics are described using voxel-wise image intensities and their spatial interaction features. To more accurately model the empirical grey level distribution of the brain signals, we use a linear combination of discrete Gaussians (LCDG) model having positive and negative components. To accurately account for the large inhomogeneity in infant MRIs, a higher-order Markov-Gibbs Random Field (MGRF) spatial interaction model that integrates third- and fourth- order families with a traditional second-order model is proposed. The proposed approach was tested and evaluated on 102 3D MR brain scans using three metrics: the Dice coefficient, the 95-percentile modified Hausdorff distance, and the absolute brain volume difference. Experimental results show better segmentation of MR brain images compared to current open source segmentation tools.

End-to-end workflow for finite element analysis of tumor treating fields in glioblastomas

Tumor Treating Fields (TTFields) therapy is an approved modality of treatment for glioblastoma. Patient anatomy-based finite element analysis (FEA) has the potential to reveal not only how these fields affect tumor control but also how to improve efficacy. While the automated tools for segmentation speed up the generation of FEA models, multi-step manual corrections are required, including removal of disconnected voxels, incorporation of unsegmented structures and the addition of 36 electrodes plus gel layers matching the TTFields transducers. Existing approaches are also not scalable for the high throughput analysis of large patient volumes. A semi-automated workflow was developed to prepare FEA models for TTFields mapping in the human brain. Magnetic resonance imaging (MRI) pre-processing, segmentation, electrode and gel placement, and post-processing were all automated. The material properties of each tissue were applied to their corresponding mask in silico using COMSOL Multiphysics (COMSOL, Burlington, MA, USA). The fidelity of the segmentations with and without post-processing was compared against the full semi-automated segmentation workflow approach using Dice coefficient analysis. The average relative differences for the electric fields generated by COMSOL were calculated in addition to observed differences in electric field-volume histograms. Furthermore, the mesh file formats in MPHTXT and NASTRAN were also compared using the differences in the electric field-volume histogram. The Dice coefficient was less for auto-segmentation without versus auto-segmentation with post-processing, indicating convergence on a manually corrected model. An existent but marginal relative difference of electric field maps from models with manual correction versus those without was identified, and a clear advantage of using the NASTRAN mesh file format was found. The software and workflow outlined in this article may be used to accelerate the investigation of TTFields in glioblastoma patients by facilitating the creation of FEA models derived from patient MRI datasets.

Fast Segmentation of Colour Apple Image under All-Weather Natural Conditions for Vision Recognition of Picking Robots

In order to resolve the poor real-time performance problem of the normalized cut (Ncut) method in apple vision recognition of picking robots, a fast segmentation method of colour apple images based on the adaptive mean-shift and Ncut methods is proposed in this paper. Firstly, the traditional Ncut method based on pixels is changed into the Ncut method based on regions by the adaptive mean-shift initial segmenting. In this way, the number of peaks and edges in the image is dramatically reduced and the computation speed is improved. Secondly, the image is divided into regional maps by extracting the R-B colour feature, which not only reduces the quantity of regions, but also to some extent overcomes the effect on illumination. On this basis, every region map is expressed by a region point, so the undirected graph of the R-B colour grey-level feature is attained. Finally, regarding the undirected graph as the input of Ncut, we construct the weight matrix W by region points and determine the number of clusters based on the decision-theoretic rough set. The adaptive clustering segmentation can be implemented by an Ncut algorithm. Experimental results show that the maximum segmentation error is 3% and the average recognition time is less than 0.7s, which can meet the requirements of a real-time picking robot.

A Five-Level Wavelet Decomposition and Dimensional Reduction Approach for Feature Extraction and Classification of MR and CT Scan Images

This paper presents a two-dimensional wavelet based decomposition algorithm for classification of biomedical images. The two-dimensional wavelet decomposition is done up to five levels for the input images. Histograms of decomposed images are then used to form the feature set. This feature set is further reduced using probabilistic principal component analysis. The reduced set of features is then fed into either K nearest neighbor algorithm or feed-forward artificial neural network, to classify images. The algorithm is compared with three other techniques in terms of accuracy. The proposed algorithm has been found better up to 3.3%, 12.75%, and 13.75% on average over the first, second, and third algorithm, respectively, using KNN and up to 6.22%, 13.9%, and 14.1% on average using ANN. The dataset used for comparison consisted of CT Scan images of lungs and MR images of heart as obtained from different sources.

Dynamic Susceptibility Contrast-MRI Quantification Software Tool: Development and Evaluation

Relative cerebral blood volume (rCBV) is a magnetic resonance imaging biomarker that is used to differentiate progression from pseudoprogression in patients with glioblastoma multiforme, the most common primary brain tumor. However, calculated rCBV depends considerably on the software used. Automating all steps required for rCBV calculation is important, as user interaction can lead to increased variability and possible inaccuracies in clinical decision-making. Here, we present an automated tool for computing rCBV from dynamic susceptibility contrast-magnetic resonance imaging that includes leakage correction. The entrance and exit bolus time points are automatically calculated using wavelet-based detection. The proposed tool is compared with 3 Food and Drug Administration-approved software packages, 1 automatic and 2 requiring user interaction, on a data set of 43 patients. We also evaluate manual and automated white matter (WM) selection for normalization of the cerebral blood volume maps. Our system showed good agreement with 2 of the 3 software packages. The intraclass correlation coefficient for all comparisons between the same software operated by different people was >0.880, except for FuncTool when operated by user 1 versus user 2. Little variability in agreement between software tools was observed when using different WM selection techniques. Our algorithm for automatic rCBV calculation with leakage correction and automated WM selection agrees well with 2 out of the 3 FDA-approved software packages.

Segmenting and validating brain tissue definitions in the presence of varying tissue contrast

We propose a method for segmenting brain tissue as either gray matter or white matter in the presence of varying tissue contrast, which can derive from either differential changes in tissue water content or increasing myelin content of white matter. Our method models the spatial distribution of intensities as a Markov Random Field (MRF) and estimates the parameters for the MRF model using a maximum likelihood approach. Although previously described methods have used similar models to segment brain tissue, accurate model of the conditional probabilities of tissue intensities and adaptive estimates of tissue properties to local intensities generates tissue definitions that are accurate and robust to variations in tissue contrast with age and across illnesses. Robustness to variations in tissue contrast is important to understand normal brain development and to identify the brain bases of neurological and psychiatric illnesses. We used simulated brains of varying tissue contrast to compare both visually and quantitatively the performance of our method with the performance of prior methods. We assessed validity of the cortical definitions by associating cortical thickness with various demographic features, clinical measures, and medication use in our three large cohorts of participants who were either healthy or who had Bipolar Disorder (BD), Autism Spectrum Disorder (ASD), or familial risk for Major Depressive Disorder (MDD). We assessed validity of the tissue definitions using synthetic brains and data for three large cohort of individuals with various neuropsychiatric disorders. Visual inspection and quantitative analyses showed that our method accurately and robustly defined the cortical mantle in brain images with varying contrast. Furthermore, associating the thickness with various demographic and clinical measures generated findings that were novel and supported by histological analyses or were supported by previous MRI studies, thereby validating the cortical definitions generated by the proposed method: (1) Although cortical thickness decreased with age in adolescents, in adults cortical thickness did not correlate significantly with age. Our synthetic data showed that the previously reported thinning of cortex in adults is likely due to decease in tissue contrast, thereby suggesting that the method generated cortical definitions in adults that were invariant to tissue contrast. In adolescents, cortical thinning with age was preserved likely due to widespread dendritic and synaptic pruning, even though the effects of decreasing tissue contrast were minimized. (3) The method generated novel finding of both localized increases and decreases in thickness of males compared to females after controlling for the differing brain sizes, which are supported by the histological analyses of brain tissue in males and females. (4) The proposed method, unlike prior methods, defined thicker cortex in BD individuals using lithium. The novel finding is supported by the studies that showed lithium treatment increased dendritic arborization and neurogenesis, thereby leading to thickening of cortex. (5) In both BD and ASD participants, associations of more severe symptoms with thinner cortex showed that correcting for the effects of tissue contrast preserved the biological consequences of illnesses. Therefore, consistency of the findings across the three large cohorts of participants, in images acquired on either 1.5T or 3T MRI scanners, and with findings from prior histological analyses provides strong evidence that the proposed method generated valid and accurate definitions of the cortex while controlling for the effects of tissue contrast.

Tone-Mapped Mean-Shift Based Environment Map Sampling

In this paper, we present a novel approach for environment map sampling, which is an effective and pragmatic technique to reduce the computational cost of realistic rendering and get plausible rendering images. The proposed approach exploits the advantage of adaptive mean-shift image clustering with aid of tone-mapping, yielding oversegmented strata that have uniform intensities and capture shapes of light regions. The resulted strata, however, have unbalanced importance metric values for rendering, and the strata number is not user-controlled. To handle these issues, we develop an adaptive split-and-merge scheme that refines the strata and obtains a better balanced strata distribution. Compared to the state-of-the-art methods, our approach achieves comparable and even better rendering quality in terms of SSIM, RMSE and HDRVDP2 image quality metrics. Experimental results further show that our approach is more robust to the variation of viewpoint, environment rotation, and sample number.

Temporal-spatial mean-shift clustering analysis to improve functional MRI activation detection

Cluster analysis (CA) is often used in functional magnetic resonance imaging (fMRI) analysis to improve detection of functional activations. Commonly used clustering techniques typically only consider spatial information of a statistical parametric image (SPI) in their calculations. This study examines incorporating the temporal characteristics of acquired fMRI data with mean-shift clustering (MSC) for fMRI analysis to enhance activation detections. Simulated data and real fMRI data was used to compare the commonly used cluster analysis with MSC using a feature space containing temporal characteristics. Receiver Operating Characteristic curves show that improvements in low contrast to noise scenarios using MSC over CA and our previous MSC technique at all tested simulated activation sizes. The proposed MSC technique with a feature space using both temporal and spatial data characteristics shows improved activation detection for both simulated and real Blood oxygen level dependent (BOLD) fMRI data (approximately 60% increase). The proposed techniques are useful in techniques that inherently have low contrast to noise ratios, such as non-proton imaging or high resolution BOLD fMRI.

Unsupervised Segmentation of Head Tissues from Multi-modal MR Images for EEG Source Localization

In this paper, we present and evaluate an automatic unsupervised segmentation method, hierarchical segmentation approach (HSA)-Bayesian-based adaptive mean shift (BAMS), for use in the construction of a patient-specific head conductivity model for electroencephalography (EEG) source localization. It is based on a HSA and BAMS for segmenting the tissues from multi-modal magnetic resonance (MR) head images. The evaluation of the proposed method was done both directly in terms of segmentation accuracy and indirectly in terms of source localization accuracy. The direct evaluation was performed relative to a commonly used reference method brain extraction tool (BET)-FMRIB's automated segmentation tool (FAST) and four variants of the HSA using both synthetic data and real data from ten subjects. The synthetic data includes multiple realizations of four different noise levels and several realizations of typical noise with a 20% bias field level. The Dice index and Hausdorff distance were used to measure the segmentation accuracy. The indirect evaluation was performed relative to the reference method BET-FAST using synthetic two-dimensional (2D) multimodal magnetic resonance (MR) data with 3% noise and synthetic EEG (generated for a prescribed source). The source localization accuracy was determined in terms of localization error and relative error of potential. The experimental results demonstrate the efficacy of HSA-BAMS, its robustness to noise and the bias field, and that it provides better segmentation accuracy than the reference method and variants of the HSA. They also show that it leads to a more accurate localization accuracy than the commonly used reference method and suggest that it has potential as a surrogate for expert manual segmentation for the EEG source localization problem.

Fully automated segmentation of corpus callosum in midsagittal brain MRIs

In the diagnosis of various brain disorders by analyzing the brain magnetic resonance images (MRI), the segmentation of corpus callosum (CC) is a crucial step. In this paper, we propose a fully automated technique for CC segmentation in the T1-weighted midsagittal brain MRIs. An adaptive mean shift clustering technique is first used to cluster homogenous regions in the image. In order to distinguish the CC from other brain tissues, area analysis, template matching, in conjunction with the shape and location analysis are proposed to identify the CC area. The boundary of detected CC area is then used as the initial contour in the Geometric Active Contour (GAC) model, and evolved to get the final segmentation result. Experimental results demonstrate that the proposed technique overcomes the problem of manual initialization in existing GAC technique, and provides a reliable segmentation performance.

On the fully automatic construction of a realistic head model for EEG source localization

Accurate multi-tissue segmentation of magnetic resonance (MR) images is an essential first step in the construction of a realistic finite element head conductivity model (FEHCM) for electroencephalography (EEG) source localization. All of the segmentation approaches proposed to date for this purpose require manual intervention or correction and are thus laborious, time-consuming, and subjective. In this paper we propose and evaluate a fully automatic method based on a hierarchical segmentation approach (HSA) incorporating Bayesian-based adaptive mean-shift segmentation (BAMS). An evaluation of HSA-BAMS, as well as two reference methods, in terms of both segmentation accuracy and the source localization accuracy of the resulting FEHCM is also presented. The evaluation was performed using (i) synthetic 2D multi-modal MRI head data and synthetic EEG (generated for a prescribed source), and (ii) real 3D T1-weighted MRI head data and real EEG data (with expert determined source localization). Expert manual segmentation served as segmentation ground truth. The results show that HSA-BAMS outperforms the two reference methods and that it can be used as a surrogate for manual segmentation for the construction of a realistic FEHCM for EEG source localization.

Building up a Bio-Inspired Visual Attention Model by Integrating Top-Down Shape Bias and Improved Mean Shift Adaptive Segmentation

The driver-assistance system (DAS) becomes quite necessary in-vehicle equipment nowadays due to the large number of road traffic accidents worldwide. An efficient DAS detecting hazardous situations robustly is key to reduce road accidents. The core of a DAS is to identify salient regions or regions of interest relevant to visual attended objects in real visual scenes for further process. In order to achieve this goal, we present a method to locate regions of interest automatically based on a novel adaptive mean shift segmentation algorithm to obtain saliency objects. In the proposed mean shift algorithm, we use adaptive Bayesian bandwidth to find the convergence of all data points by iterations and the k-nearest neighborhood queries. Experiments showed that the proposed algorithm is efficient, and yields better visual salient regions comparing with ground-truth benchmark. The proposed algorithm continuously outperformed other known visual saliency methods, generated higher precision and better recall rates, when challenged with natural scenes collected locally and one of the largest publicly available data sets. The proposed algorithm can also be extended naturally to detect moving vehicles in dynamic scenes once integrated with top-down shape biased cues, as demonstrated in our experiments.

Transfer learning improves supervised image segmentation across imaging protocols

The variation between images obtained with different scanners or different imaging protocols presents a major challenge in automatic segmentation of biomedical images. This variation especially hampers the application of otherwise successful supervised-learning techniques which, in order to perform well, often require a large amount of labeled training data that is exactly representative of the target data. We therefore propose to use transfer learning for image segmentation. Transfer-learning techniques can cope with differences in distributions between training and target data, and therefore may improve performance over supervised learning for segmentation across scanners and scan protocols. We present four transfer classifiers that can train a classification scheme with only a small amount of representative training data, in addition to a larger amount of other training data with slightly different characteristics. The performance of the four transfer classifiers was compared to that of standard supervised classification on two magnetic resonance imaging brain-segmentation tasks with multi-site data: white matter, gray matter, and cerebrospinal fluid segmentation; and white-matter-/MS-lesion segmentation. The experiments showed that when there is only a small amount of representative training data available, transfer learning can greatly outperform common supervised-learning approaches, minimizing classification errors by up to 60%.

Automatic segmentation of myocardium from black-blood MR images using entropy and local neighborhood information

By using entropy and local neighborhood information, we present in this study a robust adaptive Gaussian regularizing Chan-Vese (CV) model to segment the myocardium from magnetic resonance images with intensity inhomogeneity. By utilizing the circular Hough transformation (CHT) our model is able to detect epicardial and endocardial contours of the left ventricle (LV) as circles automatically, and the circles are used as the initialization. In the cost functional of our model, the interior and exterior energies are weighted by the entropy to improve the robustness of the evolving curve. Local neighborhood information is used to evolve the level set function to reduce the impact of the heterogeneity inside the regions and to improve the segmentation accuracy. An adaptive window is utilized to reduce the sensitivity to initialization. The Gaussian kernel is used to regularize the level set function, which can not only ensure the smoothness and stability of the level set function, but also eliminate the traditional Euclidean length term and re-initialization. Extensive validation of the proposed method on patient data demonstrates its superior performance over other state-of-the-art methods.

MRI segmentation of the human brain: challenges, methods, and applications

Image segmentation is one of the most important tasks in medical image analysis and is often the first and the most critical step in many clinical applications. In brain MRI analysis, image segmentation is commonly used for measuring and visualizing the brain's anatomical structures, for analyzing brain changes, for delineating pathological regions, and for surgical planning and image-guided interventions. In the last few decades, various segmentation techniques of different accuracy and degree of complexity have been developed and reported in the literature. In this paper we review the most popular methods commonly used for brain MRI segmentation. We highlight differences between them and discuss their capabilities, advantages, and limitations. To address the complexity and challenges of the brain MRI segmentation problem, we first introduce the basic concepts of image segmentation. Then, we explain different MRI preprocessing steps including image registration, bias field correction, and removal of nonbrain tissue. Finally, after reviewing different brain MRI segmentation methods, we discuss the validation problem in brain MRI segmentation.

Fully automatic lesion segmentation in breast MRI using mean-shift and graph-cuts on a region adjacency graph

PURPOSE

To present and evaluate a fully automatic method for segmentation (i.e., detection and delineation) of suspicious tissue in breast MRI.

MATERIALS AND METHODS

The method, based on mean-shift clustering and graph-cuts on a region adjacency graph, was developed and its parameters tuned using multimodal (T1, T2, DCE-MRI) clinical breast MRI data from 35 subjects (training data). It was then tested using two data sets. Test set 1 comprises data for 85 subjects (93 lesions) acquired using the same protocol and scanner system used to acquire the training data. Test set 2 comprises data for eight subjects (nine lesions) acquired using a similar protocol but a different vendor's scanner system. Each lesion was manually delineated in three-dimensions by an experienced breast radiographer to establish segmentation ground truth. The regions of interest identified by the method were compared with the ground truth and the detection and delineation accuracies quantitatively evaluated.

RESULTS

One hundred percent of the lesions were detected with a mean of 4.5 ± 1.2 false positives per subject. This false-positive rate is nearly 50% better than previously reported for a fully automatic breast lesion detection system. The median Dice coefficient for Test set 1 was 0.76 (interquartile range, 0.17), and 0.75 (interquartile range, 0.16) for Test set 2.

CONCLUSION

The results demonstrate the efficacy and accuracy of the proposed method as well as its potential for direct application across different MRI systems. It is (to the authors' knowledge) the first fully automatic method for breast lesion detection and delineation in breast MRI.

3D multimodal MRI brain glioma tumor and edema segmentation: a graph cut distribution matching approach

This study investigates a fast distribution-matching, data-driven algorithm for 3D multimodal MRI brain glioma tumor and edema segmentation in different modalities. We learn non-parametric model distributions which characterize the normal regions in the current data. Then, we state our segmentation problems as the optimization of several cost functions of the same form, each containing two terms: (i) a distribution matching prior, which evaluates a global similarity between distributions, and (ii) a smoothness prior to avoid the occurrence of small, isolated regions in the solution. Obtained following recent bound-relaxation results, the optima of the cost functions yield the complement of the tumor region or edema region in nearly real-time. Based on global rather than pixel wise information, the proposed algorithm does not require an external learning from a large, manually-segmented training set, as is the case of the existing methods. Therefore, the ensuing results are independent of the choice of a training set. Quantitative evaluations over the publicly available training and testing data set from the MICCAI multimodal brain tumor segmentation challenge (BraTS 2012) demonstrated that our algorithm yields a highly competitive performance for complete edema and tumor segmentation, among nine existing competing methods, with an interesting computing execution time (less than 0.5s per image).

Three-dimensional brain magnetic resonance imaging segmentation via knowledge-driven decision theory

Brain tissue segmentation on magnetic resonance (MR) imaging is a difficult task because of significant intensity overlap between the tissue classes. We present a new knowledge-driven decision theory (KDT) approach that incorporates prior information of the relative extents of intensity overlap between tissue class pairs for volumetric MR tissue segmentation. The proposed approach better handles intensity overlap between tissues without explicitly employing methods for removal of MR image corruptions (such as bias field). Adaptive tissue class priors are employed that combine probabilistic atlas maps with spatial contextual information obtained from Markov random fields to guide tissue segmentation. The energy function is minimized using a variational level-set-based framework, which has shown great promise for MR image analysis. We evaluate the proposed method on two well-established real MR datasets with expert ground-truth segmentations and compare our approach against existing segmentation methods. KDT has low-computational complexity and shows better segmentation performance than other segmentation methods evaluated using these MR datasets.

Application of mean-shift clustering to blood oxygen level dependent functional MRI activation detection

Background

Functional magnetic resonance imaging (fMRI) analysis is commonly done with cross-correlation analysis (CCA) and the General Linear Model (GLM). Both CCA and GLM techniques, however, typically perform calculations on a per-voxel basis and do not consider relationships neighboring voxels may have. Clustered voxel analyses have then been developed to improve fMRI signal detections by taking advantages of relationships of neighboring voxels. Mean-shift clustering (MSC) is another technique which takes into account properties of neighboring voxels and can be considered for enhancing fMRI activation detection.

Methods

This study examines the adoption of MSC to fMRI analysis. MSC was applied to a Statistical Parameter Image generated with the CCA technique on both simulated and real fMRI data. The MSC technique was then compared with CCA and CCA plus cluster analysis. A range of kernel sizes were used to examine how the technique behaves.

Results

Receiver Operating Characteristic curves shows an improvement over CCA and Cluster analysis. False positive rates are lower with the proposed technique. MSC allows the use of a low intensity threshold and also does not require the use of a cluster size threshold, which improves detection of weak activations and highly focused activations.

Conclusion

The proposed technique shows improved activation detection for both simulated and real Blood Oxygen Level Dependent fMRI data. More detailed studies are required to further develop the proposed technique.

Adaptive segmentation of vertebral bodies from sagittal MR images based on local spatial information and Gaussian weighted chi-square distance

We present a novel method for the automatic segmentation of the vertebral bodies from 2D sagittal magnetic resonance (MR) images of the spine. First, a new affinity matrix is constructed by incorporating neighboring information, which local intensity is considered to depict the image and overcome the noise effectively. Second, the Gaussian kernel function is to weight chi-square distance based on the neighboring information, which the vital spatial structure of the image is introduced to improve the accuracy of the segmentation task. Third, an adaptive local scaling parameter is utilized to facilitate the image segmentation and avoid the optimal configuration of controlling parameter manually. The encouraging results on the spinal MR images demonstrate the advantage of the proposed method over other methods in terms of both efficiency and robustness.

Hierarchical statistical shape models of multiobject anatomical structures: application to brain MRI

The accurate segmentation of subcortical brain structures in magnetic resonance (MR) images is of crucial importance in the interdisciplinary field of medical imaging. Although statistical approaches such as active shape models (ASMs) have proven to be particularly useful in the modeling of multiobject shapes, they are inefficient when facing challenging problems. Based on the wavelet transform, the fully generic multiresolution framework presented in this paper allows us to decompose the interobject relationships into different levels of detail. The aim of this hierarchical decomposition is twofold: to efficiently characterize the relationships between objects and their particular localities. Experiments performed on an eight-object structure defined in axial cross sectional MR brain images show that the new hierarchical segmentation significantly improves the accuracy of the segmentation, and while it exhibits a remarkable robustness with respect to the size of the training set.

A framework for analysis of brain cine MR sequences

In this paper, we propose a framework to automate the assessment of the movements of a third cerebral ventricle in a cine MR sequence. Indeed, the goal of this assessment is to build an atlas of the movements of the healthy ventricles in the context of the hydrocephalus pathology. This approach is composed of two phases: a contour extraction, using fractional integration and a registration method, based on dynamic evolutionary optimization. The first phase of the framework is based on the fractional integration thresholding, that allows delineating the contours of the area of interest. In order to track over time each point of the primitive and achieve the assessment of the deformation, a matching method, based on a new dynamic optimization algorithm, called Dynamic Covariance Matrix Adaptation Evolution Strategy (D-CMAES), is used. The obtained results for quantification have been clinically validated by an expert and compared to those presented in the literature.