Automated segmentation of brain MR images

A simple, robust and efficient image segmentation algorithm for classifying brain tissues from dual echo Magnetic Resonance (MR) images is presented. The algorithm consists of a sequence of adaptive histogram analysis, morphological operations and knowledge based rules to accurately classify various regions such as the brain matter and the cerebrospinal fluid, and detect if there are any abnormal regions. It can be completely automated and has been tested on over hundred images from several patient studies. Experimental results are provided.     

Automatic MR image segmentation using maximization of mutual information

Microsystem Technologies - Magnetic resonance (MR) brain image segmentation is an important task for the early detection of any deformation followed by the quantitative analysis for the prediction...     

Automatic segmentation of lung fields on chest radiographic images.

In this work we have implemented a system for the automatic segmentation of lung fields in chest radiographic images. The image analysis process is carried out in three levels. In the first one we perform operations on the image that are independent from domain knowledge. This knowledge is implicitly and not very elaborately used in the intermediate level and used in an explicit manner in the high level block, globally corresponding to the idea of progressive segmentation. The representation of knowledge in the high level block is in the form of production rules. The control structure is in general bottom-up but there are certain hybrid control stages, in which the control is driven by the region model (main organs) we are seeking. We have applied the global system to a set of 45 posteroanterior (PA) chest radiographs, obtaining a mean degree of overlap with contours drawn by radiologists of 87%.     

A framework with modified fast FCM for brain MR images segmentation

Intensity inhomogeneity, noise and partial volume (PV) effect render a challenging task for segmentation of brain magnetic resonance (MR) images. Most of the current MR image segmentation methods focus on only one or two of the effects listed above. In this paper, a framework with modified fast fuzzy c-means for brain MR images segmentation is proposed to take all these effects into account simultaneously and improve the accuracy of image segmentations. Firstly, we propose a new automated method to determine the initial values of the centroids. Secondly, an adaptive method to incorporate the local spatial continuity is proposed to overcome the noise effectively and prevent the edge from blurring. The intensity inhomogeneity is estimated by a linear combination of a set of basis functions. Meanwhile, a regularization term is added to reduce the iteration steps and accelerate the algorithm. The weights of the regularization terms are all automatically computed to avoid the manually tuned parameter. Synthetic and real MR images are used to test the proposed framework. Improved performance of the proposed algorithm is observed where the intensity inhomogeneity, noise and PV effect are commonly encountered. The experimental results show that the proposed method has stronger anti-noise property and higher segmentation precision than other reported FCM-based techniques.     

Iterative sliced inverse regression for segmentation of ultrasound and MR images

In this study, we propose an integrated approach based on iterative sliced inverse regression (ISIR) for the segmentation of ultrasound and magnetic resonance (MR) images. The approach integrates two stages. The first is the unsupervised clustering which combines multidimensional scaling (MDS) with K-Means. The dimension reduction based on MDS is employed to obtain fewer representative variates as input variables for K-Means. This step intends to generate the initial group labels of the training data for the second stage of supervised segmentation. We then combine the SIR with the nearest mean classifier (NMC) or the support vector machine (SVM) to iteratively update the group labels for supervised segmentation. The method of SIR is introduced by Li [Sliced inverse regression for dimension reduction. J. Am. Stat. Assoc. 86 (1991) 316–342] to explore the effective dimension reduction (e.d.r.) directions from the training data embedded in high-dimensional space. The test data are then projected onto these directions and the classifiers are further applied to classify the test data. The integrated approach based on ISIR is evaluated on simulated and clinical images, which include ultrasound and MR images. The evaluation results indicate that this approach provides an improvement of image segmentation over the methods to be compared without dimension reduction.     

Automatic joint classification and segmentation of whole cell 3D images

We present a machine learning tool for automatic texton-based joint classification and segmentation of mitochondria in MNT-1 cells imaged using ion-abrasion scanning electron microscopy (IA-SEM). For diagnosing signatures that may be unique to cellular states such as cancer, automatic tools with minimal user intervention need to be developed for analysis and mining of high-throughput data from these large volume data sets (typically ). Challenges for such a tool in 3D electron microscopy arise due to low contrast and signal-to-noise ratios (SNR) inherent to biological imaging. Our approach is based on block-wise classification of images into a trained list of regions. Given manually labeled images, our goal is to learn models that can localize novel instances of the regions in test datasets. Since datasets obtained using electron microscopes are intrinsically noisy, we improve the SNR of the data for automatic segmentation by implementing a 2D texture-preserving filter on each slice of the 3D dataset. We investigate texton-based region features in this work. Classification is performed by k-nearest neighbor (k-NN) classifier, support vector machines (SVMs), adaptive boosting (AdaBoost) and histogram matching using a NN classifier. In addition, we study the computational complexity vs. segmentation accuracy tradeoff of these classifiers. Segmentation results demonstrate that our approach using minimal training data performs close to semi-automatic methods using the variational level-set method and manual segmentation carried out by an experienced user. Using our method, which we show to have minimal user intervention and high classification accuracy, we investigate quantitative parameters such as volume of the cytoplasm occupied by mitochondria, differences between the surface area of inner and outer membranes and mean mitochondrial width which are quantities potentially relevant to distinguishing cancer cells from normal cells. To test the accuracy of our approach, these quantities are compared against manually computed counterparts. We also demonstrate extension of these methods to segment 3D images obtained using electron tomography.     

A new segmentation system for brain MR images based on fuzzy techniques

This work concerns a new method called fuzzy membership C-means (FMCMs) for segmentation of magnetic resonance images (MRI), and an efficient program implementation of it to the segmentation of MRI. Classical unsupervised clustering methods including the FCM by Bezdek, suffer many problems that can be partially treated with a proper rule to construct the initial membership matrix to clusters. This work develops a specific method to construct the initial membership matrix to clusters in order to improve the strength of the clusters. The new FMCM is tested on a set of benchmarks and then the application to the segmentation of MR images is presented and compared with the results obtained using FCM.     

Automatic brain tumor segmentation from magnetic resonance images using superpixel-based approach

Multimedia Tools and Applications - Cancer is the second leading cause of deaths worldwide, reported by World Health Organization (WHO). The abnormal growth of cells, which should die at the time...     

Dense deformation field estimation for atlas-based segmentation of pathological MR brain images

Atlas registration is a recognized paradigm for the automatic segmentation of normal MR brain images. Unfortunately, atlas-based segmentation has been of limited use in presence of large space-occupying lesions. In fact, brain deformations induced by such lesions are added to normal anatomical variability and they may dramatically shift and deform anatomically or functionally important brain structures. In this work, we chose to focus on the problem of inter-subject registration of MR images with large tumors, inducing a significant shift of surrounding anatomical structures. First, a brief survey of the existing methods that have been proposed to deal with this problem is presented. This introduces the discussion about the requirements and desirable properties that we consider necessary to be fulfilled by a registration method in this context: To have a dense and smooth deformation field and a model of lesion growth, to model different deformability for some structures, to introduce more prior knowledge, and to use voxel-based features with a similarity measure robust to intensity differences. In a second part of this work, we propose a new approach that overcomes some of the main limitations of the existing techniques while complying with most of the desired requirements above. Our algorithm combines the mathematical framework for computing a variational flow proposed by Hermosillo et al. [G. Hermosillo, C. Chefd'Hotel, O. Faugeras, A variational approach to multi-modal image matching, Tech. Rep., INRIA (February 2001).] with the radial lesion growth pattern presented by Bach et al. [M. Bach Cuadra, C. Pollo, A. Bardera, O. Cuisenaire, J.-G. Villemure, J.-Ph. Thiran, Atlas-based segmentation of pathological MR brain images using a model of lesion growth, IEEE Trans. Med. Imag. 23 (10) (2004) 1301-1314.]. Results on patients with a meningioma are visually assessed and compared to those obtained with the most similar method from the state-of-the-art.     

Automatic segmentation of optic disc in retinal fundus images using semi-supervised deep learning

Diseases of the eye require manual segmentation and examination of the optic disc by ophthalmologists. Though, image segmentation using deep learning techniques is achieving remarkable results, it leverages on large-scale labeled datasets. But, in the field of medical imaging, it is challenging to acquire large labeled datasets. Hence, this article proposes a novel deep learning model to automatically segment the optic disc in retinal fundus images by using the concepts of semi-supervised learning and transfer learning. Initially, a convolutional autoencoder (CAE) is trained to automatically learn features from a large number of unlabeled fundus images available from the Kaggle’s diabetic retinopathy (DR) dataset. The autoencoder (AE) learns the features from the unlabeled images by reconstructing the input images and becomes a pre-trained network (model). After this, the pre-trained autoencoder network is converted into a segmentation network. Later, using transfer learning, the segmentation network is trained with retinal fundus images along with their corresponding optic disc ground truth images from the DRISHTI GS1 and RIM-ONE datasets. The trained segmentation network is then tested on retinal fundus images from the test set of DRISHTI GS1 and RIM-ONE datasets. The experimental results show that the proposed method performs on par with the state-of-the-art methods achieving a 0.967 and 0.902 dice score coefficient on the test set of the DRISHTI GS1 and RIM-ONE datasets respectively. The proposed method also shows that transfer learning and semi-supervised learning overcomes the barrier imposed by the large labeled dataset. The proposed segmentation model can be used in automatic retinal image processing systems for diagnosing diseases of the eye.

     

Automatic retinal layer segmentation in SD-OCT images with CSC guided by spatial characteristics

Multimedia Tools and Applications - Segmentation of retinal layers with central serious chorioretinopathy (CSC) in Spectral Domain Optical Coherence Tomography (SD-OCT) images is significant for...     

基于模糊马尔可夫场的脑部MR图像分割算法

在传统马尔可夫场模型的基础上,建立了模糊马尔可夫场模型。通过对模型的分析得出图像像素对不同类的隶属度计算公式,提出了一种高效、无监督的图像分割算法,从而实现了对脑部MR图像的精确分割。通过对模拟脑部MR图像和临床脑部MR图像分割实验,表明新算法比传统的基于马尔可夫场的图像分割算法和模糊C-均值等图像分割算法有更精确的图像分割能力。     

Automatic MR brain image segmentation using a multiseed based multiobjective clustering approach

In this paper, the automatic segmentation of a multispectral magnetic resonance image of the brain is posed as a clustering problem in the intensity space. The automatic clustering problem is thereafter modelled as solving a multiobjective optimization (MOO) problem, optimizing a set of cluster validity indices simultaneously. A multiobjective clustering technique, named MCMOClust, is used to solve this problem. MCMOClust utilizes a recently developed simulated annealing based multiobjective optimization method as the underlying optimization strategy. Each cluster is divided into several small hyperspherical subclusters and the centers of all these small sub-clusters are encoded in a string to represent the whole clustering. For assigning points to different clusters, these local sub-clusters are considered individually. For the purpose of objective function evaluation, these sub-clusters are merged appropriately to form a variable number of global clusters. Two cluster validity indices, one based on the Euclidean distance, XB-index, and another recently developed point symmetry distance based cluster validity index, Sym-index, are optimized simultaneously to automatically evolve the appropriate number of clusters present in MR brain images. A semi-supervised method is used to select a single solution from the final Pareto optimal front of MCMOClust. The present method is applied on several simulated T1-weighted, T2-weighted and proton density normal and MS lesion magnetic resonance brain images. Superiority of the present method over Fuzzy C-means, Expectation Maximization clustering algorithms and a newly developed symmetry based fuzzy genetic clustering technique (Fuzzy-VGAPS), are demonstrated quantitatively. The automatic segmentation obtained by multiseed based multiobjective clustering technique (MCMOClust) is also compared with the available ground truth information.     

Semi-supervised discriminative classification with application to tumorous tissues segmentation of MR brain images

Due to the large data size of 3D MR brain images and the blurry boundary of the pathological tissues, tumor segmentation work is difficult. This paper introduces a discriminative classification algorithm for semi-automated segmentation of brain tumorous tissues. The classifier uses interactive hints to obtain models to classify normal and tumor tissues. A non-parametric Bayesian Gaussian random field in the semi-supervised mode is implemented. Our approach uses both labeled data and a subset of unlabeled data sampling from 2D/3D images for training the model. Fast algorithm is also developed. Experiments show that our approach produces satisfactory segmentation results comparing to the manually labeled results by experts.

Automatic selection of arterial input function on dynamic contrast-enhanced MR images

Dynamic susceptibility contrast-magnetic resonance imaging (DSC-MRI) data analysis requires the knowledge of the arterial input function (AIF) to quantify the cerebral blood flow (CBF), volume (CBV) and the mean transit time (MTT). AIF can be obtained either manually or using automatic algorithms. We present a method to derive the AIF on the middle cerebral artery (MCA). The algorithm draws a region of interest (ROI) where the MCA is located. Then, it uses a recursive cluster analysis on the ROI to select the arterial voxels. The algorithm had been compared on simulated data to literature state of art automatic algorithms and on clinical data to the manual procedure. On in silico data, our method allows to reconstruct the true AIF and it is less affected by partial volume effect bias than the other methods. In clinical data, automatic AIF provides CBF and MTT maps with a greater contrast level compared to manual AIF ones. Therefore, AIF obtained with the proposed method improves the estimate reliability and provides a quantitatively reliable physiological picture.     

Superpixel-based brain tumor segmentation in MR images using an extended local fuzzy active contour model

Multimedia Tools and Applications - In this paper, to deal with poor boundaries in the presence of noise and heterogeneity of magnetic resonance (MR) images, a new region-based fuzzy active contour...     

Interactive segmentation of the cerebral lobes with fuzzy inference in 3T MR images.

Measurement of volume and surface area of the frontal, parietal, temporal and occipital lobes from magnetic resonance (MR) images shows promise as a method for use in diagnosis of dementia. This article presents a novel computer-aided system for automatically segmenting the cerebral lobes from 3T human brain MR images. Until now, the anatomical definition of cerebral lobes on the cerebral cortex is somewhat vague for use in automatic delineation of boundary lines, and there is no definition of cerebral lobes in the interior of the cerebrum. Therefore, we have developed a new method for defining cerebral lobes on the cerebral cortex and in the interior of the cerebrum. The proposed method determines the boundaries between the lobes by deforming initial surfaces. The initial surfaces are automatically determined based on user-given landmarks. They are smoothed and deformed so that the deforming boundaries run along the hourglass portion of the three-dimensional shape of the cerebrum with fuzzy rule-based active contour and surface models. The cerebrum is divided into the cerebral lobes according to the boundaries determined using this method. The reproducibility of our system with a given subject was assessed by examining the variability of volume and surface area in three healthy subjects, with measurements performed by three beginners and one expert user. The experimental results show that our system segments the cerebral lobes with high reproducibility.     

A method for liver segmentation in perfusion MR images using probabilistic atlases and viscous reconstruction

Magnetic resonance (MR) tomographic images are routinely used in diagnosis of liver pathologies. Liver segmentation is needed for these types of images. It is therefore an important requirement for later tasks such as comparison among studies of different patients, as well as studies of the same patient (including those taken during the diffusion of a contrast, as in perfusion MR imaging). However, automatic segmentation of the liver is a challenging task due to certain reasons such as the high variability of liver shapes, similar intensity values and unclear contours between the liver and surrounding organs, especially in perfusion MR images. In order to overcome these limitations, this work proposes the use of a probabilistic atlas for liver segmentation in perfusion MR images, and the combination of the information gathered with that provided by level-based segmentation methods. The process starts with an under-segmented shape that grows slice by slice using morphological techniques (namely, viscous reconstruction); the result of the closest segmented slice and the probabilistic information provided by the atlas. Experiments with a collection of manually segmented liver images are provided, including numerical evaluation using widely accepted metrics for shape comparison.     

Automatic segmentation of adherent biological cell boundaries and nuclei from brightfield microscopy images

The detection and segmentation of adherent eukaryotic cells from brightfield microscopy images represent challenging tasks in the image analysis field. This paper presents a free and open-source image analysis package which fully automates the tasks of cell detection, cell boundary segmentation, and nucleus segmentation in brightfield images. The package also performs image registration between brightfield and fluorescence images. The algorithms were evaluated on a variety of biological cell lines and compared against manual and fluorescence-based ground truths. When tested on HT1080 and HeLa cells, the cell detection step was able to correctly identify over 80% of cells, whilst the cell boundary segmentation step was able to segment over 75% of the cell body pixels, and the nucleus segmentation step was able to correctly identify nuclei in over 75% of the cells. The algorithms for cell detection and nucleus segmentation are novel to the field, whilst the cell boundary segmentation algorithm is contrast-invariant, which makes it more robust on these low-contrast images. Together, this suite of algorithms permit brightfield microscopy image processing without the need for additional fluorescence images. Finally our sephaCe application, which is available at http://www.sephace.com, provides a novel method for integrating these methods with any motorised microscope, thus facilitating the adoption of these techniques in biological research labs.     

Automatic melanoma detection and segmentation in dermoscopy images using deep RetinaNet and conditional random fields

Multimedia Tools and Applications - Melanoma is one of the major causes of death around the world and is also known as malignant skin cancer. Melanoma detection is possible at an early stage by...