Jasjit S. Suri

Shape recovery algorithms using level sets in 2-D/3-D medical imagery: a state-of-the-art review

The class of geometric deformable models, also known as level sets, has brought tremendous impact to medical imagery due to its capability of topology preservation and fast shape recovery. In an effort to facilitate a clear and full understanding of these powerful state-of-the-art applied mathematical tools, the paper is an attempt to explore these geometric methods, their implementations and integration of regularizers to improve the robustness of these topologically independent propagating curves/surfaces. The paper first presents the origination of level sets, followed by the taxonomy of level sets. We then derive the fundamental equation of curve/surface evolution and zero-level curves/surfaces. The paper then focuses on the first core class of level sets, known as level sets without regularizers. This class presents five prototypes: gradient, edge, area-minimization, curvature-dependent and application driven. The next section is devoted to second core class of level sets, known as level sets with regularizers. In this class, we present four kinds: clustering-based, Bayesian bidirectional classifier-based, shape-based and coupled constrained-based. An entire section is dedicated to optimization and quantification techniques for shape recovery when used in the level set framework. Finally, the paper concludes with 22 general merits and four demerits on level sets and the future of level sets in medical image segmentation. We present applications of level sets to complex shapes like the human cortex acquired via MRI for neurological image analysis.

A review on MR vascular image processing: skeleton versus nonskeleton approaches: part II

Vascular segmentation has recently been given much attention. This review paper has two parts. Part I of this review focused on the physics of magnetic resonance angiography (MRA) and prefiltering techniques applied to MRA. Part II of this review presents the state-of-the-art overview, status, and new achievements in vessel segmentation algorithms from MRA. The first part of this review paper is focused on the nonskeleton or direct-based techniques. Here, we present eight different techniques along with their mathematical foundations, algorithms and their pros and cons. We will also focus on the skeleton or indirect-based techniques. We will discuss three different techniques along with their mathematical foundations, algorithms and their pros and cons. This paper also includes a clinical discussion on skeleton versus nonskeleton-based segmentation techniques. Finally, we shall conclude this paper with the possible challenges, the future, and a brief summary on vascular segmentation techniques.

A review on MR vascular image processing algorithms: acquisition and prefiltering: part I

Vascular segmentation has recently been given much attention. This review paper has two parts. Part I focuses on the physics of magnetic resonance angiography (MRA) generation and prefiltering techniques applied to MRA data sets. Part II of the review focuses on the vessel segmentation algorithms. The first section of this paper introduces the five different sets of receive coils used with the MRI system for magnetic resonance angiography data acquisition. This section then presents the five different types of the most popular data acquisition techniques: time-of-flight (TOF), phase-contrast, contrast-enhanced, black-blood, T2-weighted, and T2*-weighted, along with their pros and cons. Section II of this paper focuses on prefiltering algorithms for MRA data sets. This is necessary for removing the background nonvascular structures in the MRA data sets. Finally, the paper concludes with a clinical discussion on the challenges and the future of the data acquisition and the automated filtering algorithms.

PDE and Level Sets: Algorithmic Approaches to Static and Motion Imagery

The Contributors. 1. Review of PDEs and Level Sets. 2. Level Set Extentions, Flows and Crack Propagation 3. Geometric Regularizers for Level Sets/PDE Image Processing 4. Partial Differential Equations in Image Processing. 5. Segmentation of Motion Imagery Using PDEs. 6. Motion Image Segmentation Using Deformable Models. 7. Medical Image Segmentation Using Level Sets and PDEs. 8. Subjective Surfaces. 9. The Future of PDEs and Level Sets. 10. Index Words.

White and black blood volumetric angiographic filtering: ellipsoidal scale-space approach

Prefiltering is a critical step in three-dimensional (3D) segmentation of a blood vessel and its display. This paper presents a scale-space approach for filtering white blood and black blood angiographic volumes and its implementation issues. The raw MR angiographic volume is first converted to isotropic volume followed by 3D higher order separable Gaussian derivative convolution with known scales to generate edge volume. The edge volume is then run by the directional processor at each voxel where the eigenvalues of the 3D ellipsoid are computed. The vessel score per voxel is then estimated based on these three eigenvalues which suppress the nonvasculature and background structures yielding the filtered volume. The filtered volume is ray-cast to generate the maximum intensity projection images for display. The performance of the system is evaluated by computing the mean, variance, signal-to-noise ratio (SNR), and contrast-to-noise ratio (CNR) images. The system is run over 20 patient studies from different areas of the body such as the brain, abdomen, kidney, knee, and ankle. The computer program takes around 150 s of processing time per study for a data size of 512 /spl times/ 512 /spl times/ 194, which includes the complete performance evaluation. We also compare our strategy with the recently published MR filtering algorithms by Alexander et al. (2000) and Sun et al. (1999).

A comparison of state-of-the-art diffusion imaging techniques for smoothing medical/non-medical image data

Partial differential equations (PDEs) have dominated image processing research. The three main reasons for their success are: (1) their ability to transform a segmentation modeling problem into a partial differential equation framework and their ability to embed and integrate different regularizers into these models; (2) their ability to solve PDEs in the level set framework using finite difference methods; and (3) their easy extension to a higher dimensional space. The paper is an attempt to summarize PDEs and their solutions applied to image diffusion. The paper first presents the fundamental diffusion equation. Next, the multi-channel anisotropic diffusion imaging is presented, followed by tensor non-linear anisotropic diffusion. We also present the anisotropic diffusion based on PDE and the Tukey/Huber weight function for image noise removal. The paper also covers the recent growth of image denoising using the curve evolution approach and image denoising using histogram modification based on PDE. Finally, the paper presents non-linear image denoising. Examples covering both synthetic and real world images are presented.

A note on future research in segmentation techniques applied to neurology, cardiology, mammography and pathology

In previous chapters, we saw the application of segmentation in different areas of the body, such as the brain, heart, breast and cells. We covered many different kinds of models of CVGIP1 and PR2, but with the pace at which research in segmentation is progressing, this book would be incomplete if it did not also envision the future of segmentation techniques for the above mentioned areas. Therefore, we present in this chapter the future aspects of the segmentation techniques covered in this book.

Automatic local effect of window/level on 3D scale-space ellipsoidal filtering on run-off-arteries from white blood magnetic resonance angiography

Pre-filtering is a critical step in 3D segmentation of a blood vessel and its display. This paper presents the local effect of window/level over the 3D scale-space approach for filtering the white blood angiographic volumes and its implementation issues. The raw MR angiographic volume is first converted to an isotropic volume, then the window/level is automatically adjusted slice by slice and a composite volume is generated. 3D edges are then generated using separable Gaussian derivative convolution with known scales. The edge volume is then run by the directional processor at each voxel where the eigenvalues of the 3D ellipsoid are computed. The vessel score per voxel is then estimated based on these three eigenvalues which suppress the non-vasculature and background structures, yielding the filtered volume. The filtered volume is ray-cast to generate the maximum intensity projection images for display. The performance of the system is evaluated by computing the mean, variance, SNR and CNR images. We compare the filtering results with and without the usage of the local effect of window/level over 3D scale-space ellipsoidal filtering. We show that the automatic window/level is effective in detecting small vessels which are otherwise difficult to extrapolate. The system was run over 20 patient studies from different parts of the body such as brain, abdomen, kidney, knee, and ankle. The computer program takes around 150 seconds of processing time per study for a study with a data size of 512 /spl times/ 512 /spl times/ 194, which includes complete performance evaluation.

Segmentation techniques in the quantification of multiple sclerosis lesions in MRI

Volumetry of the brain can provide fundamental information about the development and function of the normal human brain and can yield important clues for pathology in patients suffering from neurological brain disorders (see Jernigan et al. [713]). Valuable information has been gained about the pathological processes in epilepsy (see Stone et al. [714]) and Alzheimer’s disease (see Tanabe et al. [715]) from the volume measurements of various brain structures. Brain tissue in Alzheimer’s disease was compared with elderly control volunteers by using an MR-based computerized segmentation program. Semi- automated segmentation of MR brain images revealed significant brain atrophy with significant white matter hyperintensities. In many focal diseases such as Multiple Sclerosis (MS) and cancer, the total lesion volume is indicative of the overall disease burden and may be useful in the quantification and objective evaluation of therapeutic intervention in disease (see Dastidar et al. [716] and Fillippi et al. [717]). These investigators demonstrated that MRI images provide excellent quantitative MRI tissue volume measurement. Different tissues can be identified on the images, either manually or by computer-assisted means for computing the volumes. The process of identifying and isolating a given tissue is generally referred to as segmentation. Segmentation allows color-coding of different tissues for improved delineation and makes for easier visual identification of pathology. Segmentation is evaluated as being useful in radiation therapy (see Vaidyanathan et al. [718]) and for simulating sensitive procedures for interventional neurosurgery (see Dickson et al. [719]).

Advances in computer vision, graphics, image processing and pattern recognition techniques for MR brain cortical segmentation and reconstruction: a review toward functional MRI (fMRI)

The importance of 2-D and 3-D brain segmentation has increased tremendously due to the recent growth in functional MRI (fMRI), perfusion-weighted imaging, diffusion-weighted imaging, volume graphics, 3-D segmentation, neurosurgical planning, navigation and MR brain scanning techniques. Besides that, recent growth in supervised and non-supervised brain segmentation techniques in 2-D (see Suri [322], Zavaljevski et al. [323], Barra et al. [324]) and 3-D (see Salle et al. [325], Kiebel et al. [326], Zeng et al. [327], Xu et al. [606], Fischl et al. [328], Linden et al. [329], Stokking [330], Smith [331], Hurdal [332] and ter Haar et al. [333]) have brought the engineering community, in areas such as computer vision, graphics, image processing (CVGIP) and pattern recognition, closer to the medical community, such as neuro-surgeons, psychiatrists, psychologists, physiologists, oncologists, radiologists and internists. This chapter is an attempt to review state-of-the-art cortical segmentation techniques in 2-D and 3-D using magnetic resonance imaging (MRI), and their applications. New challenges in this area are also discussed.