Automatic tumor segmentation using knowledge-based techniques

Quantitative imaging in cancer evolution and ecology

Cancer therapy, even when highly targeted, typically fails because of the remarkable capacity of malignant cells to evolve effective adaptations. These evolutionary dynamics are both a cause and a consequence of cancer system heterogeneity at many scales, ranging from genetic properties of individual cells to large-scale imaging features. Tumors of the same organ and cell type can have remarkably diverse appearances in different patients. Furthermore, even within a single tumor, marked variations in imaging features, such as necrosis or contrast enhancement, are common. Similar spatial variations recently have been reported in genetic profiles. Radiologic heterogeneity within tumors is usually governed by variations in blood flow, whereas genetic heterogeneity is typically ascribed to random mutations. However, evolution within tumors, as in all living systems, is subject to Darwinian principles; thus, it is governed by predictable and reproducible interactions between environmental selection forces and cell phenotype (not genotype). This link between regional variations in environmental properties and cellular adaptive strategies may permit clinical imaging to be used to assess and monitor intratumoral evolution in individual patients. This approach is enabled by new methods that extract, report, and analyze quantitative, reproducible, and mineable clinical imaging data. However, most current quantitative metrics lack spatialness, expressing quantitative radiologic features as a single value for a region of interest encompassing the whole tumor. In contrast, spatially explicit image analysis recognizes that tumors are heterogeneous but not well mixed and defines regionally distinct habitats, some of which appear to harbor tumor populations that are more aggressive and less treatable than others. By identifying regional variations in key environmental selection forces and evidence of cellular adaptation, clinical imaging can enable us to define intratumoral Darwinian dynamics before and during therapy. Advances in image analysis will place clinical imaging in an increasingly central role in the development of evolution-based patient-specific cancer therapy.

GBM volumetry using the 3D Slicer medical image computing platform

Volumetric change in glioblastoma multiforme (GBM) over time is a critical factor in treatment decisions. Typically, the tumor volume is computed on a slice-by-slice basis using MRI scans obtained at regular intervals. (3D)Slicer – a free platform for biomedical research – provides an alternative to this manual slice-by-slice segmentation process, which is significantly faster and requires less user interaction. In this study, 4 physicians segmented GBMs in 10 patients, once using the competitive region-growing based GrowCut segmentation module of Slicer, and once purely by drawing boundaries completely manually on a slice-by-slice basis. Furthermore, we provide a variability analysis for three physicians for 12 GBMs. The time required for GrowCut segmentation was on an average 61% of the time required for a pure manual segmentation. A comparison of Slicer-based segmentation with manual slice-by-slice segmentation resulted in a Dice Similarity Coefficient of 88.43 ± 5.23% and a Hausdorff Distance of 2.32 ± 5.23 mm.

Cerebral blood volume analysis in glioblastomas using dynamic susceptibility contrast-enhanced perfusion MRI: a comparison of manual and semiautomatic segmentation methods

Purpose

To compare the reproducibilities of manual and semiautomatic segmentation method for the measurement of normalized cerebral blood volume (nCBV) using dynamic susceptibility contrast-enhanced (DSC) perfusion MR imaging in glioblastomas.

Materials and Methods

Twenty-two patients (11 male, 11 female; 27 tumors) with histologically confirmed glioblastoma (WHO grade IV) were examined with conventional MR imaging and DSC imaging at 3T before surgery or biopsy. Then nCBV (means and standard deviations) in each mass was measured using two DSC MR perfusion analysis methods including manual and semiautomatic segmentation method, in which contrast-enhanced (CE)-T1WI and T2WI were used as structural imaging. Intraobserver and interobserver reproducibility were assessed according to each perfusion analysis method or each structural imaging. Interclass correlation coefficient (ICC), Bland-Altman plot, and coefficient of variation (CV) were used to evaluate reproducibility.

Results

Intraobserver reproducibilities on CE-T1WI and T2WI were ICC of 0.74–0.89 and CV of 20.39–36.83% in manual segmentation method, and ICC of 0.95–0.99 and CV of 8.53–16.19% in semiautomatic segmentation method, repectively. Interobserver reproducibilites on CE-T1WI and T2WI were ICC of 0.86–0.94 and CV of 19.67–35.15% in manual segmentation method, and ICC of 0.74–1.0 and CV of 5.48–49.38% in semiautomatic segmentation method, respectively. Bland-Altman plots showed a good correlation with ICC or CV in each method. The semiautomatic segmentation method showed higher intraobserver and interobserver reproducibilities at CE-T1WI-based study than other methods.

Conclusion

The best reproducibility was found using the semiautomatic segmentation method based on CE-T1WI for structural imaging in the measurement of the nCBV of glioblastomas.

Stacked autoencoders for unsupervised feature learning and multiple organ detection in a pilot study using 4D patient data

Medical image analysis remains a challenging application area for artificial intelligence. When applying machine learning, obtaining ground-truth labels for supervised learning is more difficult than in many more common applications of machine learning. This is especially so for datasets with abnormalities, as tissue types and the shapes of the organs in these datasets differ widely. However, organ detection in such an abnormal dataset may have many promising potential real-world applications, such as automatic diagnosis, automated radiotherapy planning, and medical image retrieval, where new multimodal medical images provide more information about the imaged tissues for diagnosis. Here, we test the application of deep learning methods to organ identification in magnetic resonance medical images, with visual and temporal hierarchical features learned to categorize object classes from an unlabeled multimodal DCE-MRI dataset so that only a weakly supervised training is required for a classifier. A probabilistic patch-based method was employed for multiple organ detection, with the features learned from the deep learning model. This shows the potential of the deep learning model for application to medical images, despite the difficulty of obtaining libraries of correctly labeled training datasets and despite the intrinsic abnormalities present in patient datasets.

State of the art survey on MRI brain tumor segmentation

Brain tumor segmentation consists of separating the different tumor tissues (solid or active tumor, edema, and necrosis) from normal brain tissues: gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF). In brain tumor studies, the existence of abnormal tissues may be easily detectable most of the time. However, accurate and reproducible segmentation and characterization of abnormalities are not straightforward. In the past, many researchers in the field of medical imaging and soft computing have made significant survey in the field of brain tumor segmentation. Both semiautomatic and fully automatic methods have been proposed. Clinical acceptance of segmentation techniques has depended on the simplicity of the segmentation, and the degree of user supervision. Interactive or semiautomatic methods are likely to remain dominant in practice for some time, especially in these applications where erroneous interpretations are unacceptable. This article presents an overview of the most relevant brain tumor segmentation methods, conducted after the acquisition of the image. Given the advantages of magnetic resonance imaging over other diagnostic imaging, this survey is focused on MRI brain tumor segmentation. Semiautomatic and fully automatic techniques are emphasized.

A survey of MRI-based medical image analysis for brain tumor studies

MRI-based medical image analysis for brain tumor studies is gaining attention in recent times due to an increased need for efficient and objective evaluation of large amounts of data. While the pioneering approaches applying automated methods for the analysis of brain tumor images date back almost two decades, the current methods are becoming more mature and coming closer to routine clinical application. This review aims to provide a comprehensive overview by giving a brief introduction to brain tumors and imaging of brain tumors first. Then, we review the state of the art in segmentation, registration and modeling related to tumor-bearing brain images with a focus on gliomas. The objective in the segmentation is outlining the tumor including its sub-compartments and surrounding tissues, while the main challenge in registration and modeling is the handling of morphological changes caused by the tumor. The qualities of different approaches are discussed with a focus on methods that can be applied on standard clinical imaging protocols. Finally, a critical assessment of the current state is performed and future developments and trends are addressed, giving special attention to recent developments in radiological tumor assessment guidelines.

Segmentation of pituitary adenoma: a graph-based method vs. a balloon inflation method

Among all abnormal growths inside the skull, the percentage of tumors in sellar region is approximately 10-15%, and the pituitary adenoma is the most common sellar lesion. A time-consuming process that can be shortened by using adequate algorithms is the manual segmentation of pituitary adenomas. In this contribution, two methods for pituitary adenoma segmentation in the human brain are presented and compared using magnetic resonance imaging (MRI) patient data from the clinical routine: Method A is a graph-based method that sets up a directed and weighted graph and performs a min-cut for optimal segmentation results: Method B is a balloon inflation method that uses balloon inflation forces to detect the pituitary adenoma boundaries. The ground truth of the pituitary adenoma boundaries - for the evaluation of the methods - are manually extracted by neurosurgeons. Comparison is done using the Dice Similarity Coefficient (DSC), a measure for spatial overlap of different segmentation results. The average DSC for all data sets is 77.5±4.5% for the graph-based method and 75.9±7.2% for the balloon inflation method showing no significant difference. The overall segmentation time of the implemented approaches was less than 4s - compared with a manual segmentation that took, on the average, 3.9±0.5min.

Computer assisted diagnostic system in tumor radiography

An improved and efficient method is presented in this paper to achieve a better trade-off between noise removal and edge preservation, thereby detecting the tumor region of MRI brain images automatically. Compass operator has been used in the fourth order Partial Differential Equation (PDE) based denoising technique to preserve the anatomically significant information at the edges. A new morphological technique is also introduced for stripping skull region from the brain images, which consequently leading to the process of detecting tumor accurately. Finally, automatic seeded region growing segmentation based on an improved single seed point selection algorithm is applied to detect the tumor. The method is tested on publicly available MRI brain images and it gives an average PSNR (Peak Signal to Noise Ratio) of 36.49. The obtained results also show detection accuracy of 99.46%, which is a significant improvement than that of the existing results.

A 3-D visualization method for image-guided brain surgery

This paper deals with a 3D methodology for brain tumor image-guided surgery. The methodology is based on development of a visualization process that mimics the human surgeon behavior and decision-making. In particular, it originally constructs a 3D representation of a tumor by using the segmented version of the 2D MRI images. Then it develops an optimal path for the tumor extraction based on minimizing the surgical effort and penetration area. A cost function, incorporated in this process, minimizes the damage surrounding healthy tissues taking into consideration the constraints of a new snake-like surgical tool proposed here. The tumor extraction method presented in this paper is compared with the ordinary method used on brain surgery, which is based on a straight-line based surgical tool. Illustrative examples based on real simulations present the advantages of the 3D methodology proposed here.

GLISTR: glioma image segmentation and registration

We present a generative approach for simultaneously registering a probabilistic atlas of a healthy population to brain magnetic resonance (MR) scans showing glioma and segmenting the scans into tumor as well as healthy tissue labels. The proposed method is based on the expectation maximization (EM) algorithm that incorporates a glioma growth model for atlas seeding, a process which modifies the original atlas into one with tumor and edema adapted to best match a given set of patient's images. The modified atlas is registered into the patient space and utilized for estimating the posterior probabilities of various tissue labels. EM iteratively refines the estimates of the posterior probabilities of tissue labels, the deformation field and the tumor growth model parameters. Hence, in addition to segmentation, the proposed method results in atlas registration and a low-dimensional description of the patient scans through estimation of tumor model parameters. We validate the method by automatically segmenting 10 MR scans and comparing the results to those produced by clinical experts and two state-of-the-art methods. The resulting segmentations of tumor and edema outperform the results of the reference methods, and achieve a similar accuracy from a second human rater. We additionally apply the method to 122 patients scans and report the estimated tumor model parameters and their relations with segmentation and registration results. Based on the results from this patient population, we construct a statistical atlas of the glioma by inverting the estimated deformation fields to warp the tumor segmentations of patients scans into a common space.

Neuroimaging of structural pathology and connectomics in traumatic brain injury: Toward personalized outcome prediction

Recent contributions to the body of knowledge on traumatic brain injury (TBI) favor the view that multimodal neuroimaging using structural and functional magnetic resonance imaging (MRI and fMRI, respectively) as well as diffusion tensor imaging (DTI) has excellent potential to identify novel biomarkers and predictors of TBI outcome. This is particularly the case when such methods are appropriately combined with volumetric/morphometric analysis of brain structures and with the exploration of TBI-related changes in brain network properties at the level of the connectome. In this context, our present review summarizes recent developments on the roles of these two techniques in the search for novel structural neuroimaging biomarkers that have TBI outcome prognostication value. The themes being explored cover notable trends in this area of research, including (1) the role of advanced MRI processing methods in the analysis of structural pathology, (2) the use of brain connectomics and network analysis to identify outcome biomarkers, and (3) the application of multivariate statistics to predict outcome using neuroimaging metrics. The goal of the review is to draw the community's attention to these recent advances on TBI outcome prediction methods and to encourage the development of new methodologies whereby structural neuroimaging can be used to identify biomarkers of TBI outcome.

Semi-automatic segmentation software for quantitative clinical brain glioblastoma evaluation

RATIONALE AND OBJECTIVES

Quantitative measurement provides essential information about disease progression and treatment response in patients with glioblastoma multiforme (GBM). The goal of this article is to present and validate a software pipeline for semi-automatic GBM segmentation, called AFINITI (Assisted Follow-up in NeuroImaging of Therapeutic Intervention), using clinical data from GBM patients.

MATERIALS AND METHODS

Our software adopts the current state-of-the-art tumor segmentation algorithms and combines them into one clinically usable pipeline. Both the advantages of the traditional voxel-based and the deformable shape-based segmentation are embedded into the software pipeline. The former provides an automatic tumor segmentation scheme based on T1- and T2-weighted magnetic resonance (MR) brain data, and the latter refines the segmentation results with minimal manual input.

RESULTS

Twenty-six clinical MR brain images of GBM patients were processed and compared with manual results. The results can be visualized using the embedded graphic user interface.

CONCLUSION

Validation results using clinical GBM data showed high correlation between the AFINITI results and manual annotation. Compared to the voxel-wise segmentation, AFINITI yielded more accurate results in segmenting the enhanced GBM from multimodality MR imaging data. The proposed pipeline could be used as additional information to interpret MR brain images in neuroradiology.

Template-cut: a pattern-based segmentation paradigm

We present a scale-invariant, template-based segmentation paradigm that sets up a graph and performs a graph cut to separate an object from the background. Typically graph-based schemes distribute the nodes of the graph uniformly and equidistantly on the image, and use a regularizer to bias the cut towards a particular shape. The strategy of uniform and equidistant nodes does not allow the cut to prefer more complex structures, especially when areas of the object are indistinguishable from the background. We propose a solution by introducing the concept of a "template shape" of the target object in which the nodes are sampled non-uniformly and non-equidistantly on the image. We evaluate it on 2D-images where the object's textures and backgrounds are similar, and large areas of the object have the same gray level appearance as the background. We also evaluate it in 3D on 60 brain tumor datasets for neurosurgical planning purposes.

A Patient-Specific Segmentation Framework for Longitudinal MR Images of Traumatic Brain Injury

Traumatic brain injury (TBI) is a major cause of death and disability worldwide. Robust, reproducible segmentations of MR images with TBI are crucial for quantitative analysis of recovery and treatment efficacy. However, this is a significant challenge due to severe anatomy changes caused by edema (swelling), bleeding, tissue deformation, skull fracture, and other effects related to head injury. In this paper, we introduce a multi-modal image segmentation framework for longitudinal TBI images. The framework is initialized through manual input of primary lesion sites at each time point, which are then refined by a joint approach composed of Bayesian segmentation and construction of a personalized atlas. The personalized atlas construction estimates the average of the posteriors of the Bayesian segmentation at each time point and warps the average back to each time point to provide the updated priors for Bayesian segmentation. The difference between our approach and segmenting longitudinal images independently is that we use the information from all time points to improve the segmentations. Given a manual initialization, our framework automatically segments healthy structures (white matter, grey matter, cerebrospinal fluid) as well as different lesions such as hemorrhagic lesions and edema. Our framework can handle different sets of modalities at each time point, which provides flexibility in analyzing clinical scans. We show results on three subjects with acute baseline scans and chronic follow-up scans. The results demonstrate that joint analysis of all the points yields improved segmentation compared to independent analysis of the two time points.

HYBRID APPROACH FOR BRAIN TUMOR SEGMENTATION IN MAGNETIC RESONANCE IMAGES USING CELLULAR NEURAL NETWORKS AND OPTIMIZATION TECHNIQUES

Tumor segmentation from brain magnetic resonance image data is an important but time consuming task performed manually by medical experts. Automating this process is challenging due to the high diversity in appearance of tumor tissue among different patients and in many cases, similarity between tumor and normal tissue. This paper deals with an efficient segmentation algorithm for extracting brain tumors in magnetic resonance images using Cellular Neural Networks (CNN). Learning CNN templates values are formulated as an optimization problem. The template coefficients (weights) of an CNN which will give a desired performance, can be derived by learning genetic algorithm and simulated annealing optimization techniques. The objective of this work is to compare the performance of genetic algorithm (GA) and simulated annealing (SA) for finding the optimum template values in the CNN which is used for segmenting the tumor region in the abnormal MR images. The method is applied on real data of MRI images of thirty patients with four different types of tumors. The results are compared with radiologist labeled ground truth. Quantitative analysis between ground truth and segmented tumor is presented in terms of segmentation efficiency. From the analysis and performance measures like segmentation accuracy, it is inferred that the brain tumor segmentation is best done using CNN with genetic algorithm template optimization than CNN with simulated annealing template optimization. An average accuracy rate of above 95% was obtained using this segmentation algorithm.

Uterine segmentation and volume measurement in uterine fibroid patients' MRI using fuzzy C-mean algorithm and morphological operations

BACKGROUND

Uterine fibroids are common benign tumors of the female pelvis. Uterine artery embolization (UAE) is an effective treatment of symptomatic uterine fibroids by shrinkage of the size of these tumors. Segmentation of the uterine region is essential for an accurate treatment strategy.

OBJECTIVES

In this paper, we will introduce a new method for uterine segmentation in T1W and enhanced T1W magnetic resonance (MR) images in a group of fibroid patients candidated for UAE in order to make a reliable tool for uterine volumetry.

PATIENTS AND METHODS

Uterine was initially segmented using Fuzzy C-Mean (FCM) method in T1W-enhanced images and some morphological operations were then applied to refine the initial segmentation. Finally redundant parts were removed by masking the segmented region in T1W-enhanced image over the registered T1W image and using histogram thresholding. This method was evaluated using a dataset with ten patients' images (sagittal, axial and coronal views).

RESULTS

We compared manually segmented images with the output of our system and obtained a mean similarity of 80%, mean sensitivity of 75.32% and a mean specificity of 89.5%. The Pearson correlation coefficient between the areas measured by the manual method and the automated method was 0.99.

CONCLUSIONS

The quantitative results illustrate good performance of this method. By uterine segmentation, fibroids in the uterine may be segmented and their properties may be analyzed.

Automatic segmentation of meningioma from non-contrasted brain MRI integrating fuzzy clustering and region growing

Background

In recent years, magnetic resonance imaging (MRI) has become important in brain tumor diagnosis. Using this modality, physicians can locate specific pathologies by analyzing differences in tissue character presented in different types of MR images.

This paper uses an algorithm integrating fuzzy-c-mean (FCM) and region growing techniques for automated tumor image segmentation from patients with menigioma. Only non-contrasted T1 and T2 -weighted MR images are included in the analysis. The study's aims are to correctly locate tumors in the images, and to detect those situated in the midline position of the brain.

Methods

The study used non-contrasted T1- and T2-weighted MR images from 29 patients with menigioma. After FCM clustering, 32 groups of images from each patient group were put through the region-growing procedure for pixels aggregation. Later, using knowledge-based information, the system selected tumor-containing images from these groups and merged them into one tumor image. An alternative semi-supervised method was added at this stage for comparison with the automatic method. Finally, the tumor image was optimized by a morphology operator. Results from automatic segmentation were compared to the "ground truth" (GT) on a pixel level. Overall data were then evaluated using a quantified system.

Results

The quantified parameters, including the "percent match" (PM) and "correlation ratio" (CR), suggested a high match between GT and the present study's system, as well as a fair level of correspondence. The results were compatible with those from other related studies. The system successfully detected all of the tumors situated at the midline of brain.

Six cases failed in the automatic group. One also failed in the semi-supervised alternative. The remaining five cases presented noticeable edema inside the brain. In the 23 successful cases, the PM and CR values in the two groups were highly related.

Conclusions

Results indicated that, even when using only two sets of non-contrasted MR images, the system is a reliable and efficient method of brain-tumor detection. With further development the system demonstrates high potential for practical clinical use.

Longitudinal volume analysis from computed tomography: Reproducibility using adrenal glands as surrogate tumors

This study aims to determine the precision (reproducibility) of volume assessment in routine clinical computed tomography (CT) using adrenal glands as surrogate tumors. Seven patients at our institution were identified retrospectively as having received numerous abdominal CT scans (average 13.1, range 5 to 20). The adrenal glands were used as surrogate tumors, assuming no actual volume change. Left and right adrenal gland volumes were assessed by hand segmentation for each patient scan. Over 1240 regions of interest were outlined in total. The reproducibility, expressed as the coefficient of variation (COV), was used to characterize measurement precision. The average volumes were 5.9 and 4.5 cm³ for the left and right adrenal gland, respectively, with COVs of 17.8% and 18.9%, respectively. Using one patient’s data (20 scans) as an example surrogate for a spherical tumor, it was calculated that a 13% change in volume (4.2% change in diameter) could be determined with statistical significance at P=0.05. For this case, cursor positioning error in linear measurement of object size, by even 1 pixel on the CT image, corresponded to a significant change in volume (P=0.05). The precision of volume determination was dependent on total volume. Precision improved with increasing object size (r² =0.367). Given the small dimensions of the adrenal glands, the ~18% COV is likely to be a high estimate compared to larger tumors. Modern CT scanners working with thinner sections (i.e. <1 mm) are likely to produce better measurement precision. The use of volume measurement to quantify changing tumor size is supported as a more precise metric than linear measurement.

Quantification of SPIO nanoparticles in vivo using the finite perturber method

The susceptibility gradients generated by super-paramagnetic iron oxide (SPIO) nanoparticles make them an ideal contrast agent in magnetic resonance imaging. Traditional quantification methods for SPIO nanoparticle-based contrast agents rely on either mapping T₂* values within a region or by modeling the magnetic field inhomogeneities generated by the contrast agent. In this study, a new model-based SPIO quantification method is introduced. The proposed method models magnetic field inhomogeneities by approximating regions containing SPIOs as ensembles of magnetic dipoles, referred to as the finite perturber method. The proposed method was verified using data acquired from a phantom and in vivo mouse models. The phantom consisted of an agar solution with four embedded vials, each vial containing known but different concentrations of SPIO nanoparticles. Gaussian noise was also added to the phantom data to test performance of the proposed method. The in vivo dataset was acquired using five mice, each of which was subcutaneously implanted in the flanks with 1 × 10(5) labeled and 1 × 10(6) unlabeled C6 glioma cells. For the phantom data set, the proposed algorithm was generate accurate estimations of the concentration of SPIOs. For the in vivo dataset, the method was able to give estimations of the concentration within SPIO-labeled tumors that are reasonably close to the known concentration.

Monitoring radiographic brain tumor progression

Determining radiographic progression in primary malignant brain tumors has posed a significant challenge to the neuroncology community. Glioblastoma multiforme (GBM, WHO Grade IV) through its inherent heterogeneous enhancement, growth patterns, and irregular nature has been difficult to assess for progression. Our ability to detect tumor progression radiographically remains inadequate. Despite the advanced imaging techniques, detecting tumor progression continues to be a clinical challenge. Here we review the different criteria used to detect tumor progression, and highlight the inherent challenges with detection of progression.

A medical software system for volumetric analysis of cerebral pathologies in magnetic resonance imaging (MRI) data

In this contribution, a medical software system for volumetric analysis of different cerebral pathologies in magnetic resonance imaging (MRI) data is presented. The software system is based on a semi-automatic segmentation algorithm and helps to overcome the time-consuming process of volume determination during monitoring of a patient. After imaging, the parameter settings-including a seed point-are set up in the system and an automatic segmentation is performed by a novel graph-based approach. Manually reviewing the result leads to reseeding, adding seed points or an automatic surface mesh generation. The mesh is saved for monitoring the patient and for comparisons with follow-up scans. Based on the mesh, the system performs a voxelization and volume calculation, which leads to diagnosis and therefore further treatment decisions. The overall system has been tested with different cerebral pathologies-glioblastoma multiforme, pituitary adenomas and cerebral aneurysms- and evaluated against manual expert segmentations using the Dice Similarity Coefficient (DSC). Additionally, intra-physician segmentations have been performed to provide a quality measure for the presented system.

A novel method for volumetric MRI response assessment of enhancing brain tumors

Current radiographic response criteria for brain tumors have difficulty describing changes surrounding postoperative resection cavities. Volumetric techniques may offer improved assessment, however usually are time-consuming, subjective and require expert opinion and specialized magnetic resonance imaging (MRI) sequences. We describe the application of a novel volumetric software algorithm that is nearly fully automated and uses standard T1 pre- and post-contrast MRI sequences. T1-weighted pre- and post-contrast images are automatically fused and normalized. The tumor region of interest is grossly outlined by the user. An atlas of the nasal mucosa is automatically detected and used to normalize levels of enhancement. The volume of enhancing tumor is then automatically calculated. We tested the ability of our method to calculate enhancing tumor volume with resection cavity collapse and when the enhancing tumor is obscured by subacute blood in a resection cavity. To determine variability in results, we compared narrowly-defined tumor regions with tumor regions that include adjacent meningeal enhancement and also compared different contrast enhancement threshold levels used for the automatic calculation of enhancing tumor volume. Our method quantified enhancing tumor volume despite resection cavity collapse. It detected tumor volume increase in the midst of blood products that incorrectly caused decreased measurements by other techniques. Similar trends in volume changes across scans were seen with inclusion or exclusion of meningeal enhancement and despite different automated thresholds for tissue enhancement. Our approach appears to overcome many of the challenges with response assessment of enhancing brain tumors and warrants further examination and validation.

Multivariate image segmentation using semantic region growing with adaptive edge penalty

Multivariate image segmentation is a challenging task, influenced by large intraclass variation that reduces class distinguishability as well as increased feature space sparseness and solution space complexity that impose computational cost and degrade algorithmic robustness. To deal with these problems, a Markov random field (MRF) based multivariate segmentation algorithm called "multivariate iterative region growing using semantics" (MIRGS) is presented. In MIRGS, the impact of intraclass variation and computational cost are reduced using the MRF spatial context model incorporated with adaptive edge penalty and applied to regions. Semantic region growing starting from watershed over-segmentation and performed alternatively with segmentation gradually reduces the solution space size, which improves segmentation effectiveness. As a multivariate iterative algorithm, MIRGS is highly sensitive to initial conditions. To suppress initialization sensitivity, it employs a region-level k -means (RKM) based initialization method, which consistently provides accurate initial conditions at low computational cost. Experiments show the superiority of RKM relative to two commonly used initialization methods. Segmentation tests on a variety of synthetic and natural multivariate images demonstrate that MIRGS consistently outperforms three other published algorithms.

Effective fuzzy c-means based kernel function in segmenting medical images

The objective of this paper is to develop an effective robust fuzzy c-means for a segmentation of breast and brain magnetic resonance images. The widely used conventional fuzzy c-means for medical image segmentations has limitations because of its squared-norm distance measure to measure the similarity between centers and data objects of medical images which are corrupted by heavy noise, outliers, and other imaging artifacts. To overcome the limitations this paper develops a novel objective function based standard objective function of fuzzy c-means that incorporates the robust kernel-induced distance for clustering the corrupted dataset of breast and brain medical images. By minimizing the novel objective function this paper obtains effective equation for optimal cluster centers and equation to achieve optimal membership grades for partitioning the given dataset. In order to solve the problems of clustering performance affected by initial centers of clusters, this paper introduces a specialized center initialization method for executing the proposed algorithm in segmenting medical images. Experiments are performed with synthetic, real breast and brain images to assess the performance of the proposed method. Further the validity of clustering results is obtained using silhouette method and this paper compares the results with the results of other recent reported fuzzy c-means methods. The experimental results show the superiority of the proposed clustering results.

Automated classification of multi-spectral MR images using Linear Discriminant Analysis

Magnetic resonance imaging (MRI) is a valuable instrument in medical science owing to its capabilities in soft tissue characterization and 3D visualization. A potential application of MRI in clinical practice is brain parenchyma classification. This work proposes a novel approach called "Unsupervised Linear Discriminant Analysis (ULDA)" to classify and segment the three major tissues, i.e. gray matter (GM), white matter (WM) and cerebral spinal fluid (CSF), from a multi-spectral MR image of the human brain. The ULDA comprises two processes, namely Target Generation Process (TGP) and Linear Discriminant Analysis (LDA) classification. TGP is a fuzzy-set process that generates a set of potential targets from unknown information, and applies these targets to train the optimal division boundary by LDA, such that three tissues GM, WM and CSF are separated. Finally, two sets of images, namely computer-generated phantom images and real MR images are used in the experiments to evaluate the effectiveness of ULDA. Experiment results reveal that UDLA segments a multi-spectral MR image much more effectively than either FMRIB's Automated Segmentation Tool (FAST) or Fuzzy C-means (FC).

Automated brain tumor segmentation using spatial accuracy-weighted hidden Markov Random Field

A variety of algorithms have been proposed for brain tumor segmentation from multi-channel sequences, however, most of them require isotropic or pseudo-isotropic resolution of the MR images. Although co-registration and interpolation of low-resolution sequences, such as T2-weighted images, onto the space of the high-resolution image, such as T1-weighted image, can be performed prior to the segmentation, the results are usually limited by partial volume effects due to interpolation of low-resolution images. To improve the quality of tumor segmentation in clinical applications where low-resolution sequences are commonly used together with high-resolution images, we propose the algorithm based on Spatial accuracy-weighted Hidden Markov random field and Expectation maximization (SHE) approach for both automated tumor and enhanced-tumor segmentation. SHE incorporates the spatial interpolation accuracy of low-resolution images into the optimization procedure of the Hidden Markov Random Field (HMRF) to segment tumor using multi-channel MR images with different resolutions, e.g., high-resolution T1-weighted and low-resolution T2-weighted images. In experiments, we evaluated this algorithm using a set of simulated multi-channel brain MR images with known ground-truth tissue segmentation and also applied it to a dataset of MR images obtained during clinical trials of brain tumor chemotherapy. The results show that more accurate tumor segmentation results can be obtained by comparing with conventional multi-channel segmentation algorithms.

Nosologic imaging of the brain: segmentation and classification using MRI and MRSI

A new technique is presented to create nosologic images of the brain based on magnetic resonance imaging (MRI) and magnetic resonance spectroscopic imaging (MRSI). A nosologic image summarizes the presence of different tissues and lesions in a single image by color coding each voxel or pixel according to the histopathological class it is assigned to. The proposed technique applies advanced methods from image processing as well as pattern recognition to segment and classify brain tumors. First, a registered brain atlas and a subject-specific abnormal tissue prior, obtained from MRSI data, are used for the segmentation. Next, the detected abnormal tissue is classified based on supervised pattern recognition methods. Class probabilities are also calculated for the segmented abnormal region. Compared to previous approaches, the new framework is more flexible and able to better exploit spatial information leading to improved nosologic images. The combined scheme offers a new way to produce high-resolution nosologic images, representing tumor heterogeneity and class probabilities, which may help clinicians in decision making.

Simulation of brain tumors in MR images for evaluation of segmentation efficacy

Obtaining validation data and comparison metrics for segmentation of magnetic resonance images (MRI) are difficult tasks due to the lack of reliable ground truth. This problem is even more evident for images presenting pathology, which can both alter tissue appearance through infiltration and cause geometric distortions. Systems for generating synthetic images with user-defined degradation by noise and intensity inhomogeneity offer the possibility for testing and comparison of segmentation methods. Such systems do not yet offer simulation of sufficiently realistic looking pathology. This paper presents a system that combines physical and statistical modeling to generate synthetic multi-modal 3D brain MRI with tumor and edema, along with the underlying anatomical ground truth, Main emphasis is placed on simulation of the major effects known for tumor MRI, such as contrast enhancement, local distortion of healthy tissue, infiltrating edema adjacent to tumors, destruction and deformation of fiber tracts, and multi-modal MRI contrast of healthy tissue and pathology. The new method synthesizes pathology in multi-modal MRI and diffusion tensor imaging (DTI) by simulating mass effect, warping and destruction of white matter fibers, and infiltration of brain tissues by tumor cells. We generate synthetic contrast enhanced MR images by simulating the accumulation of contrast agent within the brain. The appearance of the the brain tissue and tumor in MRI is simulated by synthesizing texture images from real MR images. The proposed method is able to generate synthetic ground truth and synthesized MR images with tumor and edema that exhibit comparable segmentation challenges to real tumor MRI. Such image data sets will find use in segmentation reliability studies, comparison and validation of different segmentation methods, training and teaching, or even in evaluating standards for tumor size like the RECIST criteria (response evaluation criteria in solid tumors).

A Scalable Framework For Segmenting Magnetic Resonance Images

A fast, accurate and fully automatic method of segmenting magnetic resonance images of the human brain is introduced. The approach scales well allowing fast segmentations of fine resolution images. The approach is based on modifications of the soft clustering algorithm, fuzzy c-means, that enable it to scale to large data sets. Two types of modifications to create incremental versions of fuzzy c-means are discussed. They are much faster when compared to fuzzy c-means for medium to extremely large data sets because they work on successive subsets of the data. They are comparable in quality to application of fuzzy c-means to all of the data. The clustering algorithms coupled with inhomogeneity correction and smoothing are used to create a framework for automatically segmenting magnetic resonance images of the human brain. The framework is applied to a set of normal human brain volumes acquired from different magnetic resonance scanners using different head coils, acquisition parameters and field strengths. Results are compared to those from two widely used magnetic resonance image segmentation programs, Statistical Parametric Mapping and the FMRIB Software Library (FSL). The results are comparable to FSL while providing significant speed-up and better scalability to larger volumes of data.