Guoyu Qian

Thresholding based on variance and intensity contrast

A new thresholding criterion is formulated for segmenting small objects by exploring the knowledge about intensity contrast. It is the weighted sum of within-class variance and intensity contrast between the object and background. Theoretical bounds of the weight are given for the uniformly distributed background and object, followed by the procedure to estimate the weight from prior knowledge. Tests against two real and two synthetic images show that small objects can be extracted successfully irrespective of the complexity of background and difference in class sizes.

Fast connected-component labelling in three-dimensional binary images based on iterative recursion

We propose two new methods to label connected components based on iterative recursion: one directly labels an original binary image while the other labels the boundary voxels followed by one-pass labelling of non-boundary object voxels. The novelty of the proposed methods is a fast labelling of large datasets without stack overflow and a flexible trade-off between speed and memory. For each iterative recursion: (1) the original volume is scanned in the raster order and an initially unlabelled object voxel v is selected, (2) a sub-volume with a user-defined size is formed around the selected voxel v, (3) within this sub-volume all object voxels 26-connected to v are labelled using iterations; and (4) subsequent iterative recursions are initiated from those border object voxels of the sub-volume that are 26-connected to v. Our experiments show that the time-memory trade-off is that the decrease in the execution time by one-third requires the increase in memory size by 3 orders. This trade-off is controlled by the user by changing the size of the sub-volume. Experiments on large three-dimensional brain phantom datasets (362x432x362 voxels of 56 MB (megabytes)) show that the proposed methods are three times faster on the average (with the maximum speedup of 10) than the existing iterative methods based on label equivalences with less than 1 MB memory consumption. Moreover, our algorithms are applicable to any dimensional data and are less dependant on the geometric complexity of connected components.

Fast Talairach Transformation for Magnetic Resonance Neuroimages

Abstract: We introduce and validate the Fast Talairach Transformation (FTT). FTT is a rapid version of the Talairach transformation (TT) with the modified Talairach landmarks. Landmark identification is fully automatic and done in 3 steps: calculation of midsagittal plane, computing of anterior commissure (AC) and posterior commissure (PC) landmarks, and calculation of cortical landmarks. To perform these steps, we use fast and anatomy-based algorithms employing simple operations. FTT was validated for 215 diversified T1-weighted and spoiled gradient recalled (SPGR) MRI data sets. It calculates the landmarks and warps the Talairach-Tournoux atlas fully automatically in about 5 sec on a standard computer. The average distance errors in landmark localization are (in mm): 1.16 (AC), 1.49 (PC), 0.08 (left), 0.13 (right), 0.48 (anterior), 0.16 (posterior), 0.35 (superior), and 0.52 (inferior). Extensions to FTT by introducing additional landmarks and applying nonlinear warping against the ventricular system are addressed. Application of FTT to other brain atlases of anatomy, function, tracts, cerebrovasculature, and blood supply territories is discussed. FTT may be useful in a clinical setting and research environment: (1) when the TT is used traditionally, (2) when a global brain structure positioning with quick searching and labeling is required, (3) in urgent cases for quick image interpretation (eg, acute stroke), (4) when the difference between nonlinear and piecewise linear warping is negligible, (5) when automatic processing of a large number of cases is required, (6) as an initial atlas-scan alignment before performing nonlinear warping, and (7) as an initial atlas-guided segmentation of brain structures before further local processing.

Segmentation of brain from computed tomography head images

An algorithm to determine the human brain (gray matter (GM) and white matter (WM)) from computed tomography (CT) head volumes with large slice thickness is proposed based on thresholding and brain mask propagation. Firstly, a 2D reference image is chosen to represent the intensity characteristics of the original 3D data set. Secondly, the region of interest of the reference image is determined as the space enclosed by the skull. Fuzzy C-means clustering is employed to determine the threshold for head mask and the low threshold for brain segmentation. The high threshold is calculated as the weighted intensity average of the boundary pixels between bones and GM/WM. Based on the low and high thresholds, the CT volume is binarized, followed by finding the brain candidates through distance criterion. Finally the brain is identified through brain mask propagation using the spatial relationship of neighboring axial slices. The algorithm has been validated against one non-enhanced CT and one enhanced CT volume with pathology

Fast, accurate, and automatic extraction of the modified Talairach cortical landmarks from magnetic resonance images.

The Talairach transformation is the most prevalent way to normalize brains and is hindered by, among others things, a lack of automatic determination of cortical landmarks. An algorithm to locate the modified Talairach cortical landmarks in three steps is proposed: determination of the three planes containing the landmarks; segmentation of the planes based on range‐constrained thresholding and morphologic operations; and local refinement of the segmentation to locate the landmarks. The algorithm has been validated against 62 T1‐weighted and SPGR MR diversified data sets. For each data set, it takes less than 2 s on a Pentium 4 to extract all six landmarks. The average landmark location errors are below 0.9 mm. The algorithm is robust due to incorporation of anatomic knowledge. A low computational cost results from processing of three 2D images and employing only simple operations like thresholding, basic morphologic operations, and distance transform. Magn Reson Med 53:970–976, 2005.

Automatic testing and assessment of neuroanatomy using a digital brain atlas: method and development of computer- and mobile-based applications.

Preparation of tests and students assessment by the instructor are time consuming. We address these two tasks in neuroanatomy education by employing a digital media application with a three‐dimensional (3D), interactive, fully segmented, and labeled brain atlas. The anatomical and vascular models in the atlas are linked to Terminologia Anatomica. Because the cerebral models are fully segmented and labeled, our approach enables automatic and random atlas‐derived generation of questions to test location and naming of cerebral structures. This is done in four steps: test individualization by the instructor, test taking by the students at their convenience, automatic student assessment by the application, and communication of the individual assessment to the instructor. A computer‐based application with an interactive 3D atlas and a preliminary mobile‐based application were developed to realize this approach. The application works in two test modes: instructor and student. In the instructor mode, the instructor customizes the test by setting the scope of testing and student performance criteria, which takes a few seconds. In the student mode, the student is tested and automatically assessed. Self‐testing is also feasible at any time and pace. Our approach is automatic both with respect to test generation and student assessment. It is also objective, rapid, and customizable. We believe that this approach is novel from computer‐based, mobile‐based, and atlas‐assisted standpoints. Anat Sci Educ 2:244–252, 2009.

Automatic Segmentation of Cerebrospinal Fluid, White and Gray Matter in Unenhanced Computed Tomography Images

RATIONALE AND OBJECTIVES Although segmentation algorithms for cerebrospinal fluid (CSF), white matter (WM), and gray matter (GM) on unenhanced computed tomographic (CT) images exist, there is no complete research in this area. To take into account poor image contrast and intensity variability on CT scans, the aim of this study was to derive and validate a novel, automatic, adaptive, and robust algorithm. MATERIALS AND METHODS Unenhanced CT scans of normal subjects from two different centers were used. The algorithm developed uses adaptive thresholding, connectivity, and domain knowledge and is based on heuristics on the shape of CT histogram. The slope of the intensity histogram corresponding to the three-dimensional largest connected region in a variable CSF intensity range is tracked to determine the critical intensity, which serves as an initial classifier of CSF-WM. Thresholds of CSF, WM, and GM are then optimally derived to minimize classification overlap. Multiple, null, and erroneous classifications are resolved by applying domain knowledge. RESULTS The ground-truth regions with the minimal partial volume effect were used to evaluate segmentation results using the statistical markers. Average sensitivity, Dice index, and specificity, respectively, for the first center were 95.7%, 97.0%, and 98.6% for CSF; 96.1%, 97.3%, and 98.8% for WM; and 95.2%, 94.3%, and 92.8% for GM. The results were consistent for the second data center. CONCLUSIONS The algorithm automatically identifies CSF, WM, and GM on unenhanced CT images with high accuracy, is robust to data from different scanners, does not require any parameter setting, and takes about 5 minutes in MATLAB to process a 512 × 512 × 30 scan. The algorithm has potential use in research and clinical applications.

A Liver Segmentation Algorithm Based on Wavelets and Machine Learning

This paper introduces an automatic liver parenchyma segmentation algorithm that can delineate liver in abdominal CT images. The proposed approach consists of three main steps. Firstly, a texture analysis is applied onto input abdominal CT images to extract pixel level features. Here, two main categories of features, namely Wavelet coefficients and Haralick texture descriptors are investigated. Secondly, support vector machines (SVM) are implemented to classify the data into pixel-wised liver or non-liver. Finally, specially combined morphological operations are designed as a post processor to remove noise and to delineate the liver. Our unique contributions to liver segmentation are twofold: one is that it has been proved through experiments that wavelet features present better classification than Haralick texture descriptors when SVMs are used; the other is that the combination of morphological operations with a pixel-wised SVM classifier can delineate volumetric liver accurately. The algorithm can be used in an advanced computer-aided liver disease diagnosis and surgical planning systems. Examples of applying the algorithm on real CT data are presented with performance validation based on the automatically segmented results and that of manually segmented ones.

Stroke Suite: Cad Systems for Acute Ischemic Stroke, Hemorrhagic Stroke, and Stroke in ER

We present a suite of computer aided-diagnosis (CAD) systems for acute ischemic stroke, hemorrhagic stroke, and stroke in emergency room. A software architecture common for them is described. The acute ischemic stroke CAD system supports thrombolysis. Our approach shifts the paradigm from a 2D visual inspection of individual scans/maps to atlas-assisted quantification and simultaneous visualization of multiple 2D/3D images. The hemorrhagic stroke CAD system supports the evacuation of hemorrhage by thrombolytic treatment. It aims at progression and quantification of blood clot removal. The clot is automatically segmented from CT time series, its volume measured, and displayed in 3D along with a catheter. A stroke CAD in emergency room enables rapid atlas-assisted decision support regarding the stroke and its location. Our stroke CAD systems facilitate and speed up image analysis, increase confidence of interpreters, and support decision making. They are potentially useful in diagnosis and research, particularly, for clinical trials.

Supervised grayscale thresholding based on transition regions

A new thresholding framework is proposed which is transition region based, and consists of deriving the transition region with the help of supervision and calculating the threshold from the transition region. Four ways of supervision are studied: picking up an object and a background pixel, from other clustering or segmentation results, based on sample statistics, and exploration of background proportions. The approach has been validated both quantitatively and qualitatively. It is found that the proposed approach: (1) is more robust, consistent and reliable than the conventional transition-region-based thresholding methods; and (2) is easier to implement and has wider applicability than existing supervised thresholding methods. The approach is especially useful for segmenting difficult images with multiple objects and/or serious imaging artifacts.