Image segmentation and registration of carp brain tissue slices oriented to brain atlas construction.

In the field of image analysis, geometric modeling and simulation of real-world problems, spline-based methods are powerful tools to develop smooth and accurate representation of the solution. In this dissertation, we propose several methods to improve the efficiency andaccuracy of B-spline based methods for different applications such as image registration, image segmentation and modeling neuron growth. Image registration is the process of finding accurate spatial correspondence between two or more images. This field has several applications such as feature tracking and fusion of images taken at different perspectives, time frames or even modalities. Image segmentation is the process of detecting importantfeatures from images. The image is partitioned into multiple labeled regions denoting each object of interest.The development of a B-spline based image registration framework that can capture large scale deformations through local refinement is carried out in order to achieve higher accuracy in less computational time. We present an efficient approach for Finite Element Method (FEM)-based nonrigid image registration, in which the spatial transformation is constructed using truncated hierarchical B-splines (THB-splines). Instead of uniform subdivision, we propose an adaptive local refinement scheme to only refine the areas of large change in deformation of the image. By incorporating the key advantages of THB-splinebasis functions such as linear independence, partition of unity and reduced overlap into the FEM-based framework, we improve the matrix sparsity and computational efficiency.The performance of the proposed method is demonstrated on 2D synthetic and medical images. We extend the algorithm to perform 3D nonrigid image registration suitable for large deformation and topology change. Control points are dynamically updated without constructing large matrices as in the finite element method. The proposed method isdemonstrated on 3D synthetic and medical images to show robustness with respect to topology change as compared to other image registration methods. In order to combine segmentation and registration in one framework, we present a novel approach for joint image segmentation and nonrigid registration using bidirectional composition to update the spatial transformation function. Unlike previous approaches,the implicit level set function defining the segmentation contour and the spatial transformationfunction are both represented using B-splines. This joint level set framework uses a variational form of an atlas-based segmentation together with large deformation basednonrigid registration. The improvement in the description of the segmentation result using B-splines leads to better accuracy of both the image segmentation and registration process. We propose a novel automatic neuron segmentation framework using a B-spline based activecontour deformation model with hyperelastic regularization and automatic initialization. This boundary-extraction based algorithm utilizes cubic B-splines to deform active contoursto match the neuron cell surface accurately. Using adaptive local refinement, finer level deformation of the active contour is captured using THB-splines in a multiresolution manner.By introducing hyperelastic regularization, we allow large nonlinear deformations of the active contours. Unlike other existing methods which represent neuron boundary aspiecewise constant function, we provide a more accurate and smooth representation of the neuron geometry.Lastly, we have focused on developing realistic computational models for modeling different stages of neuron growth using phase field method. The multi-resolution phase field method utilizes THB-splines to evaluate the gradient of the phase field variable andimprove smoothness. The stages modeled include lamellipodia formation, initial neurite outgrowth, axon differentiation and dendritic branching. Neuron growth is driven by the extracellular culture medium and intracellular transport of tubulin. Through comparison withexperimental observations, we can demonstrate a good reproduction of neuron morphologies at different stages of growth and allow extension towards formation of neurite networks.

Image segmentation technology is one of the important topics in the field of digital image research. However, there is no uniform standard for existing image segmentation methods, and the traditional image segmentation method is only suitable for some specific occasions. Therefore, it is very urgent to research and develop new theories and methods of image segmentation technology. Genetic algorithm is a method for calculating the optimal solution by simulating the biological evolution process in the natural selection and genetic mechanism of biological evolution. It has strong robustness, parallelism, adaptability and fast convergence. It can be applied in image segmentation technology to determine the segmentation threshold. Therefore, this paper studies the image segmentation based on genetic algorithm, and compares different image segmentation algorithms. The experimental results show that the image segmentation effect based on genetic algorithm is better than the traditional image segmentation.

We propose novel information content estimators for diffusion tensor images using binless approaches based on nearest-neighbour distances. Combining these estimators with existing tensor distance metrics allows us to generate entropy estimates that are consistent and accurate for diffusion tensor data. Further, we are able to obtain such estimators without having to reduce the dimensionality of the tensor data to the point where a binning estimator can be reliably used. We test our estimators in the context of noise estimation, image segmentation, and image registration. Results on 12 datasets from LBAM and 50 datasets from LONI show our estimators more accurately reflect the underlying DTI data and provide faster convergence rates for image segmentation and registration algorithms.

Purpose:Determine systematic deviations between 2D/3D and 3D/3D image registrations with six degrees of freedom (6DOF) for various imaging modalities and registration algorithms on the Varian Edge Linac.Methods:The 6DOF systematic errors were assessed by comparing automated 2D/3D (kV/MV vs. CT) with 3D/3D (CBCT vs. CT) image registrations from different imaging pairs, CT slice thicknesses, couch angles, similarity measures, etc., using a Rando head and a pelvic phantom. The 2D/3D image registration accuracy was evaluated at different treatment sites (intra‐cranial and extra‐cranial) by statistically analyzing 2D/3D pre‐treatment verification against 3D/3D localization of 192 Stereotactic Radiosurgery/Stereotactic Body Radiation Therapy treatment fractions for 88 patients.Results:The systematic errors of 2D/3D image registration using kV‐kV, MV‐kV and MV‐MV image pairs using 0.8 mm slice thickness CT images were within 0.3 mm and 0.3° for translations and rotations with a 95% confidence interval (CI). No significant difference between 2D/3D and 3D/3D image registrations (P>0.05) was observed for target localization at various CT slice thicknesses ranging from 0.8 to 3 mm. Couch angles (30, 45, 60 degree) did not impact the accuracy of 2D/3D image registration. Using pattern intensity with content image filtering was recommended for 2D/3D image registration to achieve the best accuracy. For the patient study, translational error was within 2 mm and rotational error was within 0.6 degrees in terms of 95% CI for 2D/3D image registration. For intra‐cranial sites, means and std. deviations of translational errors were −0.2±0.7, 0.04±0.5, 0.1±0.4 mm for LNG, LAT, VRT directions, respectively. For extra‐cranial sites, means and std. deviations of translational errors were ‐ 0.04±1, 0.2±1, 0.1±1 mm for LNG, LAT, VRT directions, respectively. 2D/3D image registration uncertainties for intra‐cranial and extra‐cranial sites were comparable.Conclusion:The Varian Edge radiosurgery 6DOF‐based system, can perform 2D/3D image registration with high accuracy for target localization in image‐guided stereotactic radiosurgery.The work was supported by a Research Scholar Grant, RSG‐15‐137‐01‐CCE from the American Cancer Society.

Mapping the Human Brain: What Is the Next Frontier?

Scalable joint segmentation and registration framework for infant brain images

Myocardial perfusion is commonly studied based on the evaluation of the left ventricular function using stress-rest gated myocardial perfusion single photon emission computed tomography (GSPECT), which provides a suitable identification of the myocardial region, facilitating the localization and characterization of perfusion abnormalities. The prevalence and clinical predictors of myocardial ischemia and infarct can be assessed from GSPECT images. Here, techniques of image analysis, namely image segmentation and registration, are integrated to automatically extract a set of features from myocardial perfusion SPECT images that are automatically classified as related to myocardial perfusion disorders or not. The solution implemented can be divided into two main parts: 1) building of a template image, segmentation of the template image and computation of its dimensions; 2) registration of the image under study with the template image previously built, extraction of the image features, statistical analysis and classification. It should be noted that the first step just needs to be performed once for a particular population. Hence, algorithms of image segmentation, registration and classification were used, specifically of k-means clustering, rigid and deformable registration and classification. The computational solution developed was tested using 180 3D images from 48 patients with healthy cardiac condition and 72 3D images from 12 patients with cardiac diseases, which were reconstructed using the filtered back projection algorithm and a low pass Butterworth filter or iterative algorithms. The images were classified into two classes: “abnormality present” and “abnormality not present”. The classification was assessed using five parameters: sensitivity, specificity, precision, accuracy and mean error rate. The results obtained shown that the solution is effective, both for female and male cardiac SPECT images that can have very different structural dimensions. Particularly, the solution demonstrated reasonable robustness against the two major difficulties in SPECT image analysis: image noise and low resolution. Furthermore, the classifier used demonstrated good specificity and accuracy, Table 1.

Joint segmentation and registration of images is a focused area of research nowadays. Jointly segmenting and registering noisy images and images having weak boundaries/intensity inhomogeneity is a challenging task. In medical image processing, joint segmentation and registration are essential methods that aid in distinguishing structures and aligning images for precise diagnosis and therapy. However, these methods encounter challenges, such as computational complexity and sensitivity to variations in image quality, which may reduce their effectiveness in real-world applications. Another major issue is still attaining effective joint segmentation and registration in the presence of artifacts or anatomical deformations. In this paper, a new nonparametric joint model is proposed for the segmentation and registration of multi-modality images having weak boundaries/noise. For segmentation purposes, the model will be utilizing local binary fitting data term and for registration, it is utilizing conditional mutual information. For regularization of the model, we are using linear curvature. The new proposed model is more efficient to segmenting and registering multi-modality images having intensity inhomogeneity, noise and/or weak boundaries. The proposed model is also tested on the images obtained from the freely available CHOAS dataset and compare the results of the proposed model with the other existing models using statistical measures such as the Jaccard similarity index, relative reduction, Dice similarity coefficient and Hausdorff distance. It can be seen that the proposed model outperforms the other existing models in terms of quantitatively and qualitatively.

Automated analysis of structural imaging such as lung Computed Tomography (CT) plays an increasingly important role in medical imaging applications. Despite significant progress in the development of image registration and segmentation methods, lung registration and segmentation remain a challenging task. In this paper, we present a novel image registration and segmentation approach, for which we develop a new mathematical formulation to jointly segment and register three-dimensional lung CT volumes. The new algorithm is based on a level-set formulation, which merges a classic Chan–Vese segmentation with the active dense displacement field estimation. Combining registration with segmentation has two key advantages: it allows to eliminate the problem of initializing surface based segmentation methods, and to incorporate prior knowledge into the registration in a mathematically justified manner, while remaining computationally attractive. We evaluate our framework on a publicly available lung CT data set to demonstrate the properties of the new formulation. The presented results show the improved accuracy for our joint segmentation and registration algorithm when compared to registration and segmentation performed separately.

In order to improve the accuracy and speed of medical image registration, a PCA based medical image feature extraction and registration algorithm is proposed. Firstly, the original image is trained and the sample matrix is generated. Then, the dimension of the sample image is reduced from the high dimensional space to the low dimensional space by using the linear transformation of principal component analysis. Finally, the image registration in low dimensional space is performed by combining the image registration algorithm of rigid transformation. Clinical experimental data show that the proposed algorithm has high speed and accuracy, which can not only significantly improve the image registration speed, but also remove the artifacts and retain the image feature information in clinical significance to the maximum extent, which provides an effective help for disease diagnosis and treatment.KeywordsPrincipal component analysisFeature extractionImage registrationSample matrixRigid transformation

Clustering-based segmentation algorithm is one of the common methods in image segmentation. However, how to avoid the influence of the initialization and the singular value is a problem we have to face. For overcoming this problem, we proposed a new image segmentation technology which is based on the AIC criterion and not affected by the singular values. The simulation results show that the method has good performance on stability and accuracy. Image segmentation is the key step in image analysis of computer vision and image processing field. It has been widely applied in various areas such as in the process of automating production, sensing images and biomedical images. The main goal of image segmentation is to simplify an image into segments that have a strong correlation with objects in the real world. The existing segmentation algorithm can be generally divided into three classes: edge detection, region- based segmentation and data clustering. Edge detection method relies on the edges of the object with background and the other objects. For the images with noise, it is difficult to separate the boundary and the target background (1). Region- based segmentation method segments an image into several meaningful sub-domains that are non-overlapping and same- nature. But it may lead to the over-segmentation, and has to combine with other methods to use (2). Data clustering method is based on the whole image and considers the distance between each data. Advantages of data clustering method are low complexity and easy to implement. Disadvantages are sensitive to the noise and selection of the initial centroids (3). Data clustering is one of the common techniques in image segmentation. Its purpose is to cluster pixels into several parts (regions) according to the feature of image. Based on the properties of clustering algorithm, researchers have proposed various image segmentation algorithm, such as based on hierarchical clustering, based on K-means clustering, based on fuzzy clustering, based on mean shift, and so on (3-5). Clustering analysis is an unsupervised process of partitioning a data set into subsets of similar data objects. The elementary principle is as far as possible to increase the difference between categories and reduce the difference within the category. In the process of image segmentation based on data clustering, thus the objective is that the similar pixels are divided into the same regions and the non-similar pixels are divided into different regions. However, Two difficulties we must face are that how to determine the cluster centers and avoid influence of noise. In this paper, a new image segmentation algorithm is proposed based on the AIC criterion. The new segmentation technology can achieve an ideal performance and not affected by the initialization of centroids and the singular values. The simulation results verify the efficiency of proposed algorithm.

Image segmentation and registration algorithm to collect thoracic skeleton semilandmarks for characterization of age and sex-based thoracic morphology variation

Automatically and accurately identifying the deformation zone of coastal slope landslides is crucial for exploring the mechanism of landslides and predicting landslide disasters. To this end, this study proposes an integrated automatic recognition method combining Image Clipping (IC), Image Information Enhancement (IE), Adaptive K-means Clustering Segmentation (AKS), and Optimization (O): IC-IE-AKS-O, which achieves precise extraction of the deformation area in coastal slope landslide images. Firstly, due to the more complex natural environment of field slopes, to extend the monitoring duration, we introduce a hierarchical operation algorithm based on the HSV color model, which effectively mitigates the impact of sunlight, rain, and foggy weather on image recognition accuracy. Secondly, this study proposes a 2D landslide image segmentation technique that combines K-means clustering with global threshold segmentation for landslide images, enabling the segmentation of small image regions with precision. Finally, we combine image information enhancement technology with image segmentation technology. To verify its effectiveness, we identify a landslide image of a coastal slope in Pingtan. The method displays an average relative error of 5.20% and 5.14% in the X and Y directions, respectively. Its advantages are threefold: (1) The combination of image information enhancement and segmentation techniques can more accurately identify landslide areas that appear blurred in the image; (2) expanding the temporal dimension of coastal slope monitoring; (3) providing excellent boundary conditions and segmentation results. The practical application of this method ensures the stable and accurate operation of the coastal slope monitoring system, providing a safeguard for the sustainable development of marine safety.

<p>In the field of medical imaging, image registration methods are useful for many applications such as inter- and intra-subject morphological comparisons, creation of population atlases, delivery of precision therapies, etc. A user may want to know which is the most suitable registration algorithm that would work best for the intended application, but the vastness of medical image registration applications makes evaluation and comparison of image registration performance a non-trivial task. In general, evaluating image registration performance is not straightforward because in most image registration applications there is an absence of “Gold Standard” or ground truth correspondence map to compare against. It is therefore the primary goal of this thesis work to provide a means for recommending the most appropriate registration algorithm for a given task. One of the contributions of this thesis is to examine image registration algorithm performance at the component level. Another contribution of this thesis is to catalog the benefits and limitations of many of the most commonly used image registration evaluation approaches. One incremental contribution of this thesis was to demonstrate how existing evaluation methods can be applied in the midpoint coordinate system to evaluate some symmetric image registration algorithms such as the SyN registration algorithm. Finally, a major contribution of this thesis was to develop tools to evaluate and visualize 2D and 3D image registration shape collapse. This thesis demonstrates that many current diffeomorphic image registration algorithms suffer from the collapse problem, provides the first visualizations of the collapse problem in 3D for simple shapes and real human brain MR images, and provides the first experiments that demonstrate how adjusting image registration parameters can mitigate the collapse problem to some extent.</p>

In clinical assisted diagnosis, it is an important way to obtain information with the help of medical images. Qualitative and quantitative analysis of brain tissue has become a research hotspot for brain diseases. Therefore, image segmentation technology is an indispensable link in medical image analysis. Due to the defects such as ambiguity, complexity, gray-scale unevenness, partial volume effect in magnetic resonance brain images, it is essential to improve the segmentation performance of classical algorithms in medical images. In this paper, multitasking and weighted fuzzy clustering algorithm are combined as a new algorithm (MT-WFCM) for MRI brain image segmentation. The proposed MT-WFCM algorithm improve the clustering performance of all tasks through common information between different magnetic resonance brain images with correlation. Besides, the difference between MT-WFCM and MT-FCM is that task weights are added to avoid negative effects between tasks in the segmentation process. According to five different comparative experiments, the MT-WFCM algorithm can mine the cooperative relationship among each task and the characteristics of each task effectively. In magnetic resonance image (MRI) segmentation, multi-task weighted fuzzy c-means clustering method can make up for the shortcomings of single-task clustering algorithm, strengthen the relationship between tasks, and get more accurate segmentation results.