Ivana Despotovic

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 brains 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.

Spatially Coherent Fuzzy Clustering for Accurate and Noise-Robust Image Segmentation

In this letter, we present a new FCM-based method for spatially coherent and noise-robust image segmentation. Our contribution is twofold: 1) the spatial information of local image features is integrated into both the similarity measure and the membership function to compensate for the effect of noise; and 2) an anisotropic neighborhood, based on phase congruency features, is introduced to allow more accurate segmentation without image smoothing. The segmentation results, for both synthetic and real images, demonstrate that our method efficiently preserves the homogeneity of the regions and is more robust to noise than related FCM-based methods.

Relationship of EEG sources of neonatal seizures to acute perinatal brain lesions seen on MRI: A pilot study

Even though it is known that neonatal seizures are associated with acute brain lesions, the relationship of electroencephalographic (EEG) seizures to acute perinatal brain lesions visible on magnetic resonance imaging (MRI) has not been objectively studied. EEG source localization is successfully used for this purpose in adults, but it has not been sufficiently explored in neonates. Therefore, we developed an integrated method for ictal EEG dipole source localization based on a realistic head model to investigate the utility of EEG source imaging in neonates with postasphyxial seizures. We describe here our method and compare the dipole seizure localization results with acute perinatal lesions seen on brain MRI in 10 full‐term infants with neonatal encephalopathy. Through experimental studies, we also explore the sensitivity of our method to the electrode positioning errors and the variations in neonatal skull geometry and conductivity. The localization results of 45 focal seizures from 10 neonates are compared with the visual analysis of EEG and MRI data, scored by expert physicians. In 9 of 10 neonates, dipole locations showed good relationship with MRI lesions and clinical data. Our experimental results also suggest that the variations in the used values for skull conductivity or thickness have little effect on the dipole localization, whereas inaccurate electrode positioning can reduce the accuracy of source estimates. The performance of our fused method indicates that ictal EEG source imaging is feasible in neonates and with further validation studies, this technique can become a useful diagnostic tool. Hum Brain Mapp 34:2402–2417, 2013.

An improved fuzzy clustering approach for image segmentation

Fuzzy clustering techniques have been widely used in automated image segmentation. However, since the standard fuzzy c-means (FCM) clustering algorithm does not consider any spatial information, it is highly sensitive to noise. In this paper, we present an extension of the FCM algorithm to overcome this drawback, by incorporating spatial neighborhood information into a new similarity measure. We consider that spatial information depends on the relative location and features of the neighboring pixels. The performance of the proposed algorithm is tested on synthetic and real images with different noise levels. Experimental quantitative and qualitative segmentation results show that the proposed method is effective, more robust to noise and preserves the homogeneity of the regions better than other FCM-based methods.

Automatic 3D graph cuts for brain cortex segmentation in patients with focal cortical dysplasia

In patients with intractable epilepsy, focal cortical dysplasia (FCD) is the most frequent malformation of cortical development. Identification of subtle FCD lesions using brain MRI scans is very often based on the cortical thickness measurement, where brain cortex segmentation is required as a preprocessing step. However, the accuracy of the selected segmentation method can highly affect the final FCD lesion detection. In this work, we propose an improved graph cuts algorithm integrating Markov random field-based energy function for more accurate brain cortex MRI segmentation. Our method uses three-label graph cuts and preforms automatic 3D MRI brain cortex segmentation integrating intensity and boundary information. The performance of the method is tested on both simulated MR brain images with different noise levels and real patients with FCD lesions. Experimental quantitative segmentation results showed that the proposed method is effective, robust to noise and achieves higher accuracy than other popular brain MRI segmentation methods. The qualitative validation, visually verified by a medical expert, showed that the FCD lesions were segmented well as a part of the cortex, indicating increased thickness and cortical deformation. The proposed technique can be successfully used in this, as well as in many other clinical applications.

Brain volume segmentation in newborn infants using multi-modal MRI with a low inter-slice resolution

Brain volume segmentation from neonatal magnetic resonance images (MRI) offers the possibility of exploring the developmental changes, measuring the brain growth, detecting early disorders and three-dimensional (3D) volume reconstruction. However, such segmentation is challenging mainly due to the fast growth process, complex anatomy of the developing brain and often poor MRI quality. Existing techniques are mainly developed for adult brain and are not applicable to neonates or require additional corrections. In this paper we present an algorithm for brain volume segmentation in neonates using T1-weighted (T1-w) and T2-weighted (T2-w) MRI with a low inter-slice resolution. The method incorporates both intensity and edge information and consists of three main steps: image pre-processing, brain segmentation and 3D brain reconstruction. Our algorithm is tested on real neonatal brain MRI with a gestational age between 39–41 weeks and achieves performance comparable to manual segmentation. Also, experimental segmentation results show that our method is effective and more accurate than segmentation methods originally developed for adults.

Development of a realistic head model for EEG event-detection and source localization in newborn infants

In this work we present an integrated method for electroencephalography (EEG) source localization in newborn infants, based on a realistic head model. To build a realistic head model we propose an interactive hybrid segmentation method for T1 magnetic resonance images (MRI), consisting of active contours, fuzzy c-means (FCM) clustering and mathematical morphology. Subsequently, we solve the localization problem using a spike train detection algorithm and an algorithm that deals with the forward and inverse problem. The performance of this fused method indicates that our realistic head model is suitable for the accurate localization of the EEG activity. We will present both initial qualitative and quantitative results.

Estimating blur at the brain gray-white matter boundary for FCD detection in MRI

Focal cortical dysplasia (FCD) is a frequent cause of epilepsy and can be detected using brain magnetic resonance imaging (MRI). One important MRI feature of FCD lesions is the blurring of the gray-white matter boundary (GWB), previously modelled by the gradient strength. However, in the absence of additional FCD descriptors, current gradient-based methods may yield false positives. Moreover, they do not explicitly quantify the level of blur which prevents from using them directly in the process of automated FCD detection. To improve the detection of FCD lesions displaying blur, we develop a novel algorithm called iterating local searches on neighborhood (ILSN). The novelty is that it measures the width of the blurry region rather than the gradient strength. The performance of our method is compared with the gradient magnitude method using precision and recall measures. The experimental results, tested on MRI data of 8 real FCD patients, indicate that our method has higher ability to correctly identify the FCD blurring than the gradient method.

Noise-Robust Method for Image Segmentation

Segmentation of noisy images is one of the most challenging problems in image analysis and any improvement of segmentation methods can highly influence the performance of many image processing applications. In automated image segmentation, the fuzzy c-means (FCM) clustering has been widely used because of its ability to model uncertainty within the data, applicability to multi-modal data and fairly robust behaviour. However, the standard FCM algorithm does not consider any information about the spatial image context and is highly sensitive to noise and other imaging artefacts. Considering above mentioned problems, we developed a new FCM-based approach for the noise-robust fuzzy clustering and we present it in this paper. In this new iterative algorithm we incorporated both spatial and feature space information into the similarity measure and the membership function. We considered that spatial information depends on the relative location and features of the neighbouring pixels. The performance of the proposed algorithm is tested on synthetic image with different noise levels and real images. Experimental quantitative and qualitative segmentation results show that our method efficiently preserves the homogeneity of the regions and is more robust to noise than other FCM-based methods.

T1- and T2-weighted spatially constrained fuzzy c-means clustering for brain MRI segmentation

The segmentation of brain tissue in magnetic resonance imaging (MRI) plays an important role in clinical analysis and is useful for many applications including studying brain diseases, surgical planning and computer assisted diagnoses. In general, accurate tissue segmentation is a difficult task, not only because of the complicated structure of the brain and the anatomical variability between subjects, but also because of the presence of noise and low tissue contrasts in the MRI images, especially in neonatal brain images. Fuzzy clustering techniques have been widely used in automated image segmentation. However, since the standard fuzzy c-means (FCM) clustering algorithm does not consider any spatial information, it is highly sensitive to noise. In this paper, we present an extension of the FCM algorithm to overcome this drawback, by combining information from both T1-weighted (T1-w) and T2-weighted (T2-w) MRI scans and by incorporating spatial information. This new spatially constrained FCM (SCFCM) clustering algorithm preserves the homogeneity of the regions better than existing FCM techniques, which often have difficulties when tissues have overlapping intensity profiles. The performance of the proposed algorithm is tested on simulated and real adult MR brain images with different noise levels, as well as on neonatal MR brain images with the gestational age of 39 weeks. Experimental quantitative and qualitative segmentation results show that the proposed method is effective and more robust to noise than other FCM-based methods. Also, SCFCM appears as a very promising tool for complex and noisy image segmentation of the neonatal brain.