Segmentation of multiple sclerosis lesions in brain MRI: A review of automated approaches

Automatic segmentation of multiple sclerosis (MS) lesions in brain MRI has been widely investigated in recent years with the goal of helping MS diagnosis and patient follow-up. However, the performance of most of the algorithms still falls far below expert expectations. In this paper, we review the main approaches to automated MS lesion segmentation. The main features of the segmentation algorithms are analysed and the most recent important techniques are classified into different strategies according to their main principle, pointing out their strengths and weaknesses and suggesting new research directions. A qualitative and quantitative comparison of the results of the approaches analysed is also presented. Finally, possible future approaches to MS lesion segmentation are discussed.     

Joint segmentation and pairing of multispectral chromosome images

Automated segmentation of touching or overlapping chromosomes in a metaphase image is a critical step for computer-aided chromosomes analysis. Conventional chromosome imaging methods acquire single-band grayscale images, and such a limitation makes the separation of touching or overlapping chromosomes challenging. In the multiplex fluorescence in situ hybridization (M-FISH) technique, each class of chromosomes can bind with a different combination of fluorophores. The M-FISH technique results in multispectral chromosome images, which has distinct spectral signatures. This paper presents a novel automated chromosome analysis method to combine the pixel-level geometric and multispectral information with decision-level pairing information. Our chromosome segmentation method uses the geometric and spectral information to partition the chromosome cluster into three regions. There will be ambiguity when combining these regions into separated chromosomes by using only spectral and geometric information. Then a graph–theoretical pairing method is introduced to resolve any remaining ambiguity of the aforementioned segmentation process. Experimental results demonstrate that the proposed joint segmentation and pairing method outperforms conventional grayscale and multispectral segmentation methods in separating touching and overlapping chromosomes.     

Automatic multiple sclerosis lesion detection in brain MRI by FLAIR thresholding

Magnetic resonance imaging (MRI) is frequently used to detect and segment multiple sclerosis lesions due to the detailed and rich information provided. We present a modified expectation-maximisation algorithm to segment brain tissues (white matter, grey matter, and cerebro-spinal fluid) as well as a partial volume class containing fluid and grey matter. This algorithm provides an initial segmentation in which lesions are not separated from tissue, thus a second step is needed to find them. This second step involves the thresholding of the FLAIR image, followed by a regionwise refinement to discard false detections. To evaluate the proposal, we used a database with 45 cases comprising 1.5T imaging data from three different hospitals with different scanner machines and with a variable lesion load per case. The results for our database point out to a higher accuracy when compared to two of the best state-of-the-art approaches.     

A hybrid of active contour model and convex hull for automated brain tumor segmentation in multimodal MRI

Multimedia Tools and Applications - Segmenting tumor automatically in human brain Magnetic Resonance (MR) images is challenging because of uneven, irregular and unstructured size and shape of the...     

News story segmentation in multiple modalities

In this paper, we describe an approach to segmenting news video based on the perceived shift in content using features spanning multiple modalities. We investigate a number of multimedia features, which serve as potential indicators of a change in story, in order to determine which are the most effective. The efficacy of our approach is demonstrated by the performance of our prototype, where a number of feature combinations demonstrate an up to 18% improvement in WindowDiff score compared to other state of the art story segmenters. In our investigation, there is no, one, clearly superior feature, rather the best segmentation occurs when there is synergy between multiple features. A further investigation into the effect on segmentation performance, while varying the number of training examples versus the number of features used, reveal that having better feature combinations is more important than having more training examples. Our work suggests that it is possible to train robust story segmenters for news video using only a handful of broadcasts, provided a good initial feature selection is made.     

Review of brain MRI image segmentation methods

Brain image segmentation is one of the most important parts of clinical diagnostic tools. Brain images mostly contain noise, inhomogeneity and sometimes deviation. Therefore, accurate segmentation of brain images is a very difficult task. However, the process of accurate segmentation of these images is very important and crucial for a correct diagnosis by clinical tools. We presented a review of the methods used in brain segmentation. The review covers imaging modalities, magnetic resonance imaging and methods for noise reduction, inhomogeneity correction and segmentation. We conclude with a discussion on the trend of future research in brain segmentation.     

Fully automated resource retrieval in telecommunications and internet converged environments

Today, most of the current retrieval approaches are targeted only to Web services, leaving out different other resource types (Telecom, IT and content). Moreover, such approaches are not aimed at ordinary users, since to find and use a service, users must specify their requests by complex expressions that represent desired services. Thus, the search and selection of relevant resources depend on the ability of the user to retrieve the most suitable ones, which is an inefficient task. In order to address these issues, in this paper we propose an innovative approach for resource retrieval over Telecommunications and Internet converged environments, which brings together traditional NLP techniques with retrieval process based on lightweight semantic approachs. The preliminary experiments show promising results in contrast to traditional approaches.     

A new automated segmentation and classification of mammogram images

Multimedia Tools and Applications - Early-stage recognition of lesions is the better probable manner for fighting against breast cancer to find a disease with the highest ratio of malignancy around...     

Scene segmentation in automated histopathology: techniques evolved from cytology automation

Image analysis techniques developed for cytology automation are shown to be applicable or adaptable to the analysis of fixed tissue sections in some relatively simple cases. More sophisticated techniques, developed as extensions of the basic methods, are suggested for more complicated cases. Techniques highlighted include: isodensity contour tracing, geometrical boundary repair, heuristic search, regional pre-processing, Hough transforms, and general parallel edge finding algorithms. Applications include: leukocyte analysis (in blood, marrow, and tissue), morphometric analysis of bone tissue, morphometry of muscle and nerve fibers, and the grading of non-Hodgkins lymphomas and of intraductal breast lesions.     

Toward breast cancer diagnosis based on automated segmentation of masses in mammograms

This work explores the use of characterization features extracted based on breast-mass contours obtained by automated segmentation methods, for the classification of masses in mammograms according to their diagnosis (benign or malignant). Two sets of mass contours were obtained via two segmentation methods (a dynamic-programming-based method and a constrained region-growing method), and simplified versions of these contours (modeling the contours as ellipses) were employed to extract a set of six features designed for characterization of mass margins (contrast between foreground region and background region, coefficient of variation of edge strength, two measures of the fuzziness of mass margins, a measure of spiculation based on relative gradient orientation, and a measure of spiculation based on edge-signature information). Three popular classifiers (Bayesian classifier, Fisher's linear discriminant, and a support vector machine) were then used to predict the diagnosis of a set of 349 masses based on each of said features and some combinations of these. The systems (each system consists of a segmentation method, a featureset, and a classifier) were compared with each other in terms of their performance on the diagnosis of the set of breast masses. It was found that, although there was a percent difference of about 14% in the average segmentation quality between methods, this was translated into an average percent difference of only 4% in the classification performance. It was also observed that the spiculation feature based on edge-signature information was distinctly better than the rest of the features, although it is not very robust to changes in the quality of the segmentation. All systems were more efficient in predicting the diagnosis of benign masses than that of the malignant masses, resulting in low sensitivity and high specificity values (e.g. 0.6 and 0.8, respectively) since the positive class in the classification experiments is the set of malignant masses. It was concluded that features extracted from automated contours can contribute to the diagnosis of breast masses in screening programs by correctly identifying a majority of benign masses.     

Fully automatic person segmentation in unconstrained video using spatio-temporal conditional random fields

The segmentation of objects and people in particular is an important problem in computer vision. In this paper, we focus on automatically segmenting a person from challenging video sequences in which we place no constraint on camera viewpoint, camera motion or the movements of a person in the scene. Our approach uses the most confident predictions from a pose detector as a form of anchor or keyframe stick figure prediction which helps guide the segmentation of other more challenging frames in the video. Since even state of the art pose detectors are unreliable on many frames –especially given that we are interested in segmentations with no camera or motion constraints –only the poses or stick figure predictions for frames with the highest confidence in a localized temporal region anchor further processing. The stick figure predictions within confident keyframes are used to extract color, position and optical flow features. Multiple conditional random fields (CRFs) are used to process blocks of video in batches, using a two dimensional CRF for detailed keyframe segmentation as well as 3D CRFs for propagating segmentations to the entire sequence of frames belonging to batches. Location information derived from the pose is also used to refine the results. Importantly, no hand labeled training data is required by our method. We discuss the use of a continuity method that reuses learnt parameters between batches of frames and show how pose predictions can also be improved by our model. We provide an extensive evaluation of our approach, comparing it with a variety of alternative grab cut based methods and a prior state of the art method. We also release our evaluation data to the community to facilitate further experiments. We find that our approach yields state of the art qualitative and quantitative performance compared to prior work and more heuristic alternative approaches.     

Computerized segmentation of whole-body bone scintigrams and its use in automated diagnostics

Bone scintigraphy or whole-body bone scan is one of the most common diagnostic procedures in nuclear medicine used in the last 25 years. Pathological conditions, technically poor image resolution and artefacts necessitate that algorithms use sufficient background knowledge of anatomy and spatial relations of bones in order to work satisfactorily. A robust knowledge based methodology for detecting reference points of the main skeletal regions that is simultaneously applied on anterior and posterior whole-body bone scintigrams is presented. Expert knowledge is represented as a set of parameterized rules which are used to support standard image-processing algorithms. Our study includes 467 consecutive, non-selected scintigrams, which is, to our knowledge the largest number of images ever used in such studies. Automatic analysis of whole-body bone scans using our segmentation algorithm gives more accurate and reliable results than previous studies. Obtained reference points are used for automatic segmentation of the skeleton, which is applied to automatic (machine learning) or manual (expert physicians) diagnostics. Preliminary experiments show that an expert system based on machine learning closely mimics the results of expert physicians.     

Classification of small lesions in dynamic breast MRI: eliminating the need for precise lesion segmentation through spatio-temporal analysis of contrast enhancement

Characterizing the dignity of breast lesions as benign or malignant is specifically difficult for small lesions; they do not exhibit typical characteristics of malignancy and are harder to segment since margins are harder to visualize. Previous attempts at using dynamic or morphologic criteria to classify small lesions (mean lesion diameter of about 1 cm) have not yielded satisfactory results. The goal of this work was to improve the classification performance in such small diagnostically challenging lesions while concurrently eliminating the need for precise lesion segmentation. To this end, we introduce a method for topological characterization of lesion enhancement patterns over time. Three Minkowski Functionals were extracted from all five post-contrast images of 60 annotated lesions on dynamic breast MRI exams. For each Minkowski Functional, topological features extracted from each post-contrast image of the lesions were combined into a high-dimensional texture feature vector. These feature vectors were classified in a machine learning task with support vector regression. For comparison, conventional Haralick texture features derived from gray-level co-occurrence matrices (GLCM) were used. A new method for extracting thresholded GLCM features was also introduced and investigated here. The best classification performance was observed with Minkowski Functionals area and perimeter, thresholded GLCM features f8 and f9, and conventional GLCM features f4 and f6. However, both Minkowski Functionals and thresholded GLCM achieved such results without lesion segmentation while the performance of GLCM features significantly deteriorated when lesions were not segmented ( $p<0.05$ ). This suggests that such advanced spatio-temporal characterization can improve the classification performance achieved in such small lesions, while simultaneously eliminating the need for precise segmentation.     

Effective fuzzy clustering techniques for segmentation of breast MRI

The goal of this work is to segment the breast into different regions, each corresponding to a different tissue, and to identify tissue regions judged abnormal, based on the signal enhancement-time information. There are a number of problems that render this task complex. Breast MRI segmentation based on the differential enhancement of image intensities can assist the clinician to detect suspicious regions. In this paper, we propose an effective segmentation method for breast contrast-enhanced MRI (ce-MRI). The segmentation method is developed based on standard fuzzy clustering techniques proposed by Bezedek. By minimizing the proposed effective objective function, this paper obtains an effective way of predicting membership grades for objects and new method to update centers. Experiments will be done with a synthetic image to show how effectively the new proposed effective fuzzy c-means (FCM) works in obtaining clusters. To show the performance of proposed FCM, this work compares the results with results of standard FCM algorithm on same synthetic image. Then the proposed method was applied to segment the clinical ce-MR images with the help of computer programing language and results have been shown visually.     

Liver segmentation in MRI images based on whale optimization algorithm

This paper proposes an approach for liver segmentation in MRI images based on Whale optimization algorithm (WOA). It is used to extract the different clusters in the abdominal image to support the segmentation process. A statistical image is prepared to define the potential liver position in the abdominal image. Then, WOA divides the image into a predefined number of clusters. The prepared statistical image is converted into a binary image and multiplied by the image clustered by WOA. This multiplication process removes a great part of other organs from the image. It is followed by some points, picked up by user interaction, representing the required clusters which reside in the area of liver. The morphological operations enhance the initial segmented liver and produces the final image. The proposed approach is tested using a set of 70 MRI images, annotated and approved by radiology specialists. The resulting image is validated using structural similarity index measure (SSIM), similarity index (SI) and other five measures. The overall accuracy of the experimental result showed accuracy of 96.75% using SSIM and 97.5 using SI%.     

Evaluating the potential of multispectral imagery to map multiple stages of tree mortality

Insect outbreaks are major forest disturbances, causing tree mortality across millions of ha in North America. Resultant spatial and temporal patterns of tree mortality can profoundly affect ecosystem structure and function. In this study, we evaluated the classification accuracy of multispectral imagery at different spatial resolutions. We used four-band digital aerial imagery (30-cm spatial resolution and aggregated to coarser resolutions) acquired over lodgepole pine-dominated stands in central Colorado recently attacked by mountain pine beetle. Classes of interest included green trees and multiple stages of post-insect attack tree mortality, including dead trees with red needles (“red-attack”), dead trees without needles (“gray-attack”), and non-forest. The 30-cm resolution image facilitated delineation of trees located in the field, which were used in image classification. We employed a maximum likelihood classifier using the green band, Red-Green Index (RGI), and Normalized Difference Vegetation Index (NDVI). Pixel-level classification accuracies using this imagery were good (overall accuracy of 87%, kappa = 0.84), although misclassification occurred between a) sunlit crowns of live (green) trees and herbaceous vegetation, and b) sunlit crowns of gray- and red-attack trees and bare soil. We explored the capability of coarser resolution imagery, aggregated from the 30-cm resolution to 1.2, 2.4, and 4.2 m, to improve classification accuracy. We found the highest accuracy at the 2.4-m resolution, where reduction in omission and commission errors and increases in overall accuracy (90%) and kappa (0.88) were achieved, and visual inspection indicated improved mapping. Pixels at this resolution included more shadow in forested regions than pixels in finer resolution imagery, thereby reducing forest canopy reflectance and allowing improved separation between forest and non-forest classes, yet were fine enough to resolve individual tree crowns better than the 4.2-m imagery. Our results illustrate that a classification of an image with a spatial resolution similar to the area of a tree crown outperforms that of finer and coarser resolution imagery for mapping tree mortality and non-forest classes. We also demonstrate that multispectral imagery can be used to separate multiple postoutbreak attack stages (i.e., red-attack and gray-attack) from other classes in the image.     

合理利用空间信息的MRI脑部图像分割

在合理利用空间信息的基础上,提出了一种更准确,紧致性和分离性更好的分割算法。该算法首先定义一个空间函数,并在其中引入一个控制参数,该参数可以对噪声点、边缘点以及区域内部的点进行区别对待,然后用空间信息更新隶属度。实验结果表明,该算法效果要明显优于sFCMpq算法及其改进算法(EsFCMpq)。     

Graph cut segmentation with a statistical shape model in cardiac MRI

Segmenting the right ventricle (RV) in magnetic resonance (MR) images is required for cardiac function assessment. The segmentation of the RV is a difficult task due to low contrast with surrounding tissues and high shape variability. To overcome these problems, we introduce a segmentation method based on a statistical shape model obtained with a principal component analysis (PCA) on a set of representative shapes of the RV. Shapes are not represented by a set of points, but by distance maps to their contour, relaxing the need for a costly landmark detection and matching process. A shape model is thus obtained by computing a PCA on the shape variations. This prior is registered onto the image via a very simple user interaction and then incorporated into the well-known graph cut framework in order to guide the segmentation. Our semi-automatic segmentation method has been applied on 248 MR images of a publicly available dataset (from MICCAI’12 Right Ventricle Segmentation Challenge). We show that encouraging results can be obtained for this challenging application.     

Motion segmentation of multiple translating objects from line correspondences

This paper presents an analytic approach to motion segmentation of multiple translating objects from line correspondences in three perspective views. The basic idea of our algorithm is to view the estimation of multiple translational motions as the estimation of a single, though more complex, multibody motion model that is then factored into the original models by polynomial differentiation. Experimental results on synthetic and real scenes are presented.