Segmentation Of Anatomical Structures Research Articles

Discriminative confidence estimation for probabilistic multi-atlas label fusion.

Quantitative neuroimaging analyses often rely on the accurate segmentation of anatomical brain structures. In contrast to manual segmentation, automatic methods offer reproducible outputs and provide scalability to study large databases. Among existing approaches, multi-atlas segmentation has recently shown to yield state-of-the-art performance in automatic segmentation of brain images. It consists in propagating the labelmaps from a set of atlases to the anatomy of a target image using image registration, and then fusing these multiple warped labelmaps into a consensus segmentation on the target image. Accurately estimating the contribution of each atlas labelmap to the final segmentation is a critical step for the success of multi-atlas segmentation. Common approaches to label fusion either rely on local patch similarity, probabilistic statistical frameworks or a combination of both. In this work, we propose a probabilistic label fusion framework based on atlas label confidences computed at each voxel of the structure of interest. Maximum likelihood atlas confidences are estimated using a supervised approach, explicitly modeling the relationship between local image appearances and segmentation errors produced by each of the atlases. We evaluate different spatial pooling strategies for modeling local segmentation errors. We also present a novel type of label-dependent appearance features based on atlas labelmaps that are used during confidence estimation to increase the accuracy of our label fusion. Our approach is evaluated on the segmentation of seven subcortical brain structures from the MICCAI 2013 SATA Challenge dataset and the hippocampi from the ADNI dataset. Overall, our results indicate that the proposed label fusion framework achieves superior performance to state-of-the-art approaches in the majority of the evaluated brain structures and shows more robustness to registration errors.

Deep learning of the sectional appearances of 3D CT images for anatomical structure segmentation based on an FCN voting method.

We propose a single network trained by pixel-to-label deep learning to address the general issue of automatic multiple organ segmentation in three-dimensional (3D) computed tomography (CT) images. Our method can be described as a voxel-wise multiple-class classification scheme for automatically assigning labels to each pixel/voxel in a 2D/3D CT image. We simplify the segmentation algorithms of anatomical structures (including multiple organs) in a CT image (generally in 3D) to a majority voting scheme over the semantic segmentation of multiple 2D slices drawn from different viewpoints with redundancy. The proposed method inherits the spirit of fully convolutional networks (FCNs) that consist of "convolution" and "deconvolution" layers for 2D semantic image segmentation, and expands the core structure with 3D-2D-3D transformations to adapt to 3D CT image segmentation. All parameters in the proposed network are trained pixel-to-label from a small number of CT cases with human annotations as the ground truth. The proposed network naturally fulfills the requirements of multiple organ segmentations in CT cases of different sizes that cover arbitrary scan regions without any adjustment. The proposed network was trained and validated using the simultaneous segmentation of 19 anatomical structures in the human torso, including 17 major organs and two special regions (lumen and content inside of stomach). Some of these structures have never been reported in previous research on CT segmentation. A database consisting of 240 (95% for training and 5% for testing) 3D CT scans, together with their manually annotated ground-truth segmentations, was used in our experiments. The results show that the 19 structures of interest were segmented with acceptable accuracy (88.1% and 87.9% voxels in the training and testing datasets, respectively, were labeled correctly) against the ground truth. We propose a single network based on pixel-to-label deep learning to address the challenging issue of anatomical structure segmentation in 3D CT cases. The novelty of this work is the policy of deep learning of the different 2D sectional appearances of 3D anatomical structures for CT cases and the majority voting of the 3D segmentation results from multiple crossed 2D sections to achieve availability and reliability with better efficiency, generality, and flexibility than conventional segmentation methods, which must be guided by human expertise.

Automatic fetal body and amniotic fluid segmentation from fetal ultrasound images by encoder-decoder network with inner layers.

This paper explores the effectiveness of applying a deep learning based method to segment the amniotic fluid and fetal tissues in fetal ultrasound (US) images. The deeply learned model firstly encodes the input image into down scaled feature maps by convolution and pooling structures, then up-scale the feature maps to confidence maps by corresponded un-pooling and convolution layers. Additional convolution layers with 1×1 sized kernels are adopted to enhance the feature representations, which could be used to further improve the discriminative learning of our model. We effectively update the weights of the network by fine-tuning on part of the layers from a pre-trained model. By conducting experiments using clinical data, the feasibility of our proposed approach is compared and discussed. The result proves that this work achieves satisfied results for segmentation of specific anatomical structures from US images.

Multi-atlas pancreas segmentation: Atlas selection based on vessel structure.

Automated organ segmentation from medical images is an indispensable component for clinical applications such as computer-aided diagnosis (CAD) and computer-assisted surgery (CAS). We utilize a multi-atlas segmentation scheme, which has recently been used in different approaches in the literature to achieve more accurate and robust segmentation of anatomical structures in computed tomography (CT) volume data. Among abdominal organs, the pancreas has large inter-patient variability in its position, size and shape. Moreover, the CT intensity of the pancreas closely resembles adjacent tissues, rendering its segmentation a challenging task. Due to this, conventional intensity-based atlas selection for pancreas segmentation often fails to select atlases that are similar in pancreas position and shape to those of the unlabeled target volume. In this paper, we propose a new atlas selection strategy based on vessel structure around the pancreatic tissue and demonstrate its application to a multi-atlas pancreas segmentation. Our method utilizes vessel structure around the pancreas to select atlases with high pancreatic resemblance to the unlabeled volume. Also, we investigate two types of applications of the vessel structure information to the atlas selection. Our segmentations were evaluated on 150 abdominal contrast-enhanced CT volumes. The experimental results showed that our approach can segment the pancreas with an average Jaccard index of 66.3% and an average Dice overlap coefficient of 78.5%.

Prenatal brain MRI samples for development of automatic segmentation, target- recognition and machine-learning algorithms to detect anatomical structures

In this data note, we present a sorted pool of fetal magnetic resonance imaging (MRI) specimens, selected for a project seeking to further develop a computer-vision software called MaZda, originally created for magnetic resonance (MR) image analysis. A link to download the samples is provided in the manuscript herein. This data descriptor further explains how and why these fetal MRI samples were selected. Firstly, thousands of cross-sectional images obtained from fetal MRI scans were processed and sorted semi-manually with other software. We did so because a built-in “samplesort” (sorting algorithm) is missing in MaZda version 5. Additionally, the software is unfortunately lacking effective and efficient algorithms to allow automatic identification and segmentation of anatomical structures in fetal MRI samples. Hence, the finals sorting steps were carried out manually via time-consuming methods — i.e. human- visual detection and classifications by the gestational age of pregnancy and the rotational plane of the MR scanner. Thus the latter correlates with the anatomical plane of the mother, rather than the hypothetical plane used to transect the fetus. In brief, we collated these fetal MRI samples in an effort to facilitate future research and discovery, especially to aid the improvement of MaZda.

Dinámica Pulmonar de Estructuras Anatómicasde Interés en Imágenes 4DCT

This paper presents an application of the Chan-Vese algorithm for a semi-automatic segmentation of anatomical structures of interest (lungs and lung tumor) in thorax 4DCT images, as well as its threedimensional reconstruction. Segmentations and reconstructions were performed in 10 CT images, which conform an inspiration-expiration cycle. The maximum displacement of the lung tumor was calculated using the reconstructions of the beginning of inspiration, beginning of expiration, and the voxel size information. The proposed method was able to succesfully segment the studied structures regardless of their size and shape. The threedimensional reconstruction allow us to visualize the dynamics of the structures of interest throughout the respiratory cycle. In the near future, we are expecting to be able to have more evidence of the good performance of the proposed segmentation approach, and to have feedback from a clinical expert, giving the fact that the knowledge of anatomical structures characteristics, such as their size and spatial location, may help in the planning of radiotherapy treatments (RT), optimizing the radiation dose to cancer cells and minimizing it in healthy organs. Therefore, the information found in this work maybe of interest for the planning of RT treatments.

Simultaneous segmentation of multiple anatomical structures on CT images using deep learning technique

Simultaneous segmentation of multiple anatomical structures on CT images using deep learning technique

Simultaneous Recognition and Segmentation of Multiple Anatomical Structures on CT Images by Using Deep Learning Approach

Simultaneous Recognition and Segmentation of Multiple Anatomical Structures on CT Images by Using Deep Learning Approach

Segmentation of anatomical structures of the heart based on echocardiography

Nowadays, many practical applications in the field of medical image processing require valid and reliable segmentation of images in the capacity of input data. Some of the commonly used imaging techniques are ultrasound, CT, and MRI. However, the main difference between the other medical imaging equipment and EchoCG is that it is safer, low cost, non-invasive and non-traumatic. Three-dimensional EchoCG is a non-invasive imaging modality that is complementary and supplementary to two-dimensional imaging and can be used to examine the cardiovascular function and anatomy in different medical settings. The challenging problems, presented by EchoCG image processing, such as speckle phenomena, noise, temporary non-stationarity of processes, unsharp boundaries, attenuation, etc. forced us to consider and compare existing methods and then to develop an innovative approach that can tackle the problems connected with clinical applications. Actual studies are related to the analysis and development of a cardiac parameters automatic detection system by EchoCG that will provide new data on the dynamics of changes in cardiac parameters and improve the accuracy and reliability of the diagnosis. Research study in image segmentation has highlighted the capabilities of image-based methods for medical applications. The focus of the research is both theoretical and practical aspects of the application of the methods. Some of the segmentation approaches can be interesting for the imaging and medical community. Performance evaluation is carried out by comparing the borders, obtained from the considered methods to those manually prescribed by a medical specialist. Promising results demonstrate the possibilities and the limitations of each technique for image segmentation problems. The developed approach allows: to eliminate errors in calculating the geometric parameters of the heart; perform the necessary conditions, such as speed, accuracy, reliability; build a master model that will be an indispensable assistant for operations on a beating heart.

Progressive multi-atlas label fusion by dictionary evolution.

Accurate segmentation of anatomical structures in medical images is important in recent imaging based studies. In the past years, multi-atlas patch-based label fusion methods have achieved a great success in medical image segmentation. In these methods, the appearance of each input image patch is first represented by an atlas patch dictionary (in the image domain), and then the latent label of the input image patch is predicted by applying the estimated representation coefficients to the corresponding anatomical labels of the atlas patches in the atlas label dictionary (in the label domain). However, due to the generally large gap between the patch appearance in the image domain and the patch structure in the label domain, the estimated (patch) representation coefficients from the image domain may not be optimal for the final label fusion, thus reducing the labeling accuracy. To address this issue, we propose a novel label fusion framework to seek for the suitable label fusion weights by progressively constructing a dynamic dictionary in a layer-by-layer manner, where the intermediate dictionaries act as a sequence of guidance to steer the transition of (patch) representation coefficients from the image domain to the label domain. Our proposed multi-layer label fusion framework is flexible enough to be applied to the existing labeling methods for improving their label fusion performance, i.e., by extending their single-layer static dictionary to the multi-layer dynamic dictionary. The experimental results show that our proposed progressive label fusion method achieves more accurate hippocampal segmentation results for the ADNI dataset, compared to the counterpart methods using only the single-layer static dictionary.

Fast and sequence-adaptive whole-brain segmentation using parametric Bayesian modeling

Quantitative analysis of magnetic resonance imaging (MRI) scans of the brain requires accurate automated segmentation of anatomical structures. A desirable feature for such segmentation methods is to be robust against changes in acquisition platform and imaging protocol. In this paper we validate the performance of a segmentation algorithm designed to meet these requirements, building upon generative parametric models previously used in tissue classification. The method is tested on four different datasets acquired with different scanners, field strengths and pulse sequences, demonstrating comparable accuracy to state-of-the-art methods on T1-weighted scans while being one to two orders of magnitude faster. The proposed algorithm is also shown to be robust against small training datasets, and readily handles images with different MRI contrast as well as multi-contrast data.

Improving Spleen Volume Estimation Via Computer-assisted Segmentation on Clinically Acquired CT Scans

Multi-atlas fusion is a promising approach for computer-assisted segmentation of anatomic structures. The purpose of this study was to evaluate the accuracy and time efficiency of multi-atlas segmentation for estimating spleen volumes on clinically acquired computed tomography (CT) scans. Under an institutional review board approval, we obtained 294 de-identified (Health Insurance Portability and Accountability Act-compliant) abdominal CT scans on 78 subjects from a recent clinical trial. We compared five pipelines for obtaining splenic volumes: Pipeline 1 - manual segmentation of all scans, Pipeline 2 - automated segmentation of all scans, Pipeline 3 - automated segmentation of all scans with manual segmentation for outliers on a rudimentary visual quality check, and Pipelines 4 and 5 - volumes derived from a unidimensional measurement of craniocaudal spleen length and three-dimensional splenic index measurements, respectively. Using Pipeline 1 results as ground truth, the accuracies of Pipelines 2-5 (Dice similarity coefficient, Pearson correlation, R-squared, and percent and absolute deviation of volume from ground truth) were compared for point estimates of splenic volume and for change in splenic volume over time. Time cost was also compared for Pipelines 1-5. Pipeline 3 was dominant in terms of both accuracy and time cost. With a Pearson correlation coefficient of 0.99, average absolute volume deviation of 23.7 cm(3), and time cost of 1 minute per scan, Pipeline 3 yielded the best results. The second-best approach was Pipeline 5, with a Pearson correlation coefficient of 0.98, absolute deviation of 46.92 cm(3), and time cost of 1 minute 30 seconds per scan. Manual segmentation (Pipeline 1) required 11 minutes per scan. A computer-automated segmentation approach with manual correction of outliers generated accurate splenic volumes with reasonable time efficiency.

Extended Modality Propagation: Image Synthesis of Pathological Cases.

This paper describes a novel generative model for the synthesis of multi-modal medical images of pathological cases based on a single label map. Our model builds upon i) a generative model commonly used for label fusion and multi-atlas patch-based segmentation of healthy anatomical structures, ii) the Modality Propagation iterative strategy used for a spatially-coherent synthesis of subject-specific scans of desired image modalities. The expression Extended Modality Propagation is coined to refer to the extension of Modality Propagation to the synthesis of images of pathological cases. Moreover, image synthesis uncertainty is estimated. An application to Magnetic Resonance Imaging synthesis of glioma-bearing brains is i) validated on the training dataset of a Multimodal Brain Tumor Image Segmentation challenge, ii) compared to the state-of-the-art in glioma image synthesis, and iii) illustrated using the output of two different tumor growth models. Such a generative model allows the generation of a large dataset of synthetic cases, which could prove useful for the training, validation, or benchmarking of image processing algorithms.

An Automatic Segmentation and Classification Framework Based on PCNN Model for Single Tooth in MicroCT Images

Accurate segmentation and classification of different anatomical structures of teeth from medical images plays an essential role in many clinical applications. Usually, the anatomical structures of teeth are manually labelled by experienced clinical doctors, which is time consuming. However, automatic segmentation and classification is a challenging task because the anatomical structures and surroundings of the tooth in medical images are rather complex. Therefore, in this paper, we propose an effective framework which is designed to segment the tooth with a Selective Binary and Gaussian Filtering Regularized Level Set (GFRLS) method improved by fully utilizing three dimensional (3D) information, and classify the tooth by employing unsupervised learning Pulse Coupled Neural Networks (PCNN) model. In order to evaluate the proposed method, the experiments are conducted on the different datasets of mandibular molars and the experimental results show that our method can achieve better accuracy and robustness compared to other four state of the art clustering methods.

Cloud-Based Evaluation of Anatomical Structure Segmentation and Landmark Detection Algorithms: VISCERAL Anatomy Benchmarks.

Variations in the shape and appearance of anatomical structures in medical images are often relevant radiological signs of disease. Automatic tools can help automate parts of this manual process. A cloud-based evaluation framework is presented in this paper including results of benchmarking current state-of-the-art medical imaging algorithms for anatomical structure segmentation and landmark detection: the VISCERAL Anatomy benchmarks. The algorithms are implemented in virtual machines in the cloud where participants can only access the training data and can be run privately by the benchmark administrators to objectively compare their performance in an unseen common test set. Overall, 120 computed tomography and magnetic resonance patient volumes were manually annotated to create a standard Gold Corpus containing a total of 1295 structures and 1760 landmarks. Ten participants contributed with automatic algorithms for the organ segmentation task, and three for the landmark localization task. Different algorithms obtained the best scores in the four available imaging modalities and for subsets of anatomical structures. The annotation framework, resulting data set, evaluation setup, results and performance analysis from the three VISCERAL Anatomy benchmarks are presented in this article. Both the VISCERAL data set and Silver Corpus generated with the fusion of the participant algorithms on a larger set of non-manually-annotated medical images are available to the research community.

A robust statistics driven volume-scalable active contour for segmenting anatomical structures in volumetric medical images with complex conditions.

BackgroundAccurate segmentation of anatomical structures in medical images is a critical step in the development of computer assisted intervention systems. However, complex image conditions, such as intensity inhomogeneity, noise and weak object boundary, often cause considerable difficulties in medical image segmentation. To cope with these difficulties, we propose a novel robust statistics driven volume-scalable active contour framework, to extract desired object boundary from magnetic resonance (MR) and computed tomography (CT) imagery in 3D.MethodsWe define an energy functional in terms of the initial seeded labels and two fitting functions that are derived from object local robust statistics features. This energy is then incorporated into a level set scheme which drives the active contour evolving and converging at the desired position of the object boundary. Due to the local robust statistics and the volume scaling function in the energy fitting term, the object features in local volumes are learned adaptively to guide the motion of the contours, which thereby guarantees the capability of our method to cope with intensity inhomogeneity, noise and weak boundary. In addition, the initialization of active contour is simplified by select several seeds in the object and/or background to eliminate the sensitivity to initialization.ResultsThe proposed method was applied to extensive public available volumetric medical images with challenging image conditions. The segmentation results of various anatomical structures, such as white matter (WM), atrium, caudate nucleus and brain tumor, were evaluated quantitatively by comparing with the corresponding ground truths. It was found that the proposed method achieves consistent and coherent segmentation accuracy of 0.9246 ± 0.0068 for WM, 0.9043 ± 0.0131 for liver tumors, 0.8725 ± 0.0374 for caudate nucleus, 0.8802 ± 0.0595 for brain tumors, etc., measured by Dice similarity coefficients value for the overlap between the algorithm one and the ground truth. Further comparative experimental results showed desirable performances of the proposed method over several well-known segmentation methods in terms of accuracy and robustness.ConclusionWe proposed an approach to accurate segment volumetric medical images with complex conditions. The accuracy of segmentation, robustness to noise and contour initialization were validated on the basis of extensive MR and CT volumes.

Segmentation Fusion Techniques with Application to Plenoptic Images: A Survey.

The segmentation of anatomical and pathological structures plays a key role in the characterization of clinically relevant evidence from digital images. Recently, plenoptic imaging has emerged as a new promise to enrich the diagnostic potential of conventional photography. Since the plenoptic images comprises a set of slightly different versions of the target scene, we propose to make use of those images to improve the segmentation quality in relation to the scenario of a single image segmentation. The problem of finding a segmentation solution from multiple images of a single scene, is called segmentation fusion. This paper reviews the issue of segmentation fusion in order to find solutions that can be applied to plenoptic images, particularly images from the ophthalmological domain.

Fast and Automatic Segmentation of Anatomical Chest Structures from CT Images Based on Image-Guided Lung Biopsy Surgery

Fast and Automatic Segmentation of Anatomical Chest Structures from CT Images Based on Image-Guided Lung Biopsy Surgery

Automation with Anatomical Intelligence as a Novel Pathway in Echocardiography for the Advancement of Measurements and Analysis

Clinical environments are rapidly changing to demand higher throughputs, lower cost, and more successful clinical outcomes. Echocardiography has, in turn, been forced to evolve. In addition to providing real-time imaging at a relatively low cost and risk to the patient, it now also provides rapid and robust measurements that aid in monitoring, diagnosis, and intervention. These advancements of measurements and analysis are achieved through automation afforded by Anatomical Intelligence, which is an automatic recognition and segmentation of anatomic structures within an echo image by a trained algorithm. Early versions of Anatomical Intelligence have focused on expediting the workflow of an expert sonographer, whereas, future versions will focus on enabling novice sonographers to perform tasks previously only achievable by experts.

Segmentation and surface reconstruction of a cadaver heart on Mimics software.

The Visible Korean research team used Mimics software (Materialise, Leuven, Belgium) for the segmentation and subsequent surface reconstruction of heart structures using information obtained from sectioned images of a cadaver. Twenty-six heart components were outlined in advance on Photoshop (Adobe Systems, San Jose, CA, USA). By use of the Mimics, the outlined images were then browsed along with the vertical planes as well as the 3-dimensional surface models, which were immediately built by piling the images. Erroneous delineation was readily detected and revised until satisfactory heart models were acquired. The surface models and the selected sectioned images in horizontal, coronal, and sagittal planes were inputted into a PDF file, where any combinations of reconstructed constituents could be displayed and rotated by the user. Mimics software accelerated the segmentation and surface reconstruction of heart anatomical structures. Similar benefits hopefully result from various serial images of other organs. The PDF file, and plane and stereoscopic image data are being distributed to others, and should prove valuable for medical students and clinicians.