Haralick's Method Research Articles

Assessment of Intratumor Heterogeneity in Parametric Dynamic Contrast-Enhanced MR Images: A Comparative Study of Novel and Established Methods.

Intratumor heterogeneity is associated with aggressive disease and poor survival rates in several types of cancer. A novel method for assessing intratumor heterogeneity in medical images, named the spatial gradient method, has been developed in our laboratory. In this study, we measure intratumor heterogeneity in K trans maps derived by dynamic contrast-enhanced magnetic resonance imaging using the spatial gradient method, and we compare the performance of the novel method with that of histogram analyses and texture analyses using the Haralick method. K trans maps of 58 untreated and sunitinib-treated pancreatic ductal adenocaricoma (PDAC) xenografts from two PDAC models were investigated. Intratumor heterogeneity parameters derived by the spatial gradient method were sensitive to tumor line differences as well as sunitinib-induced changes in intratumor heterogeneity. Furthermore, the parameters provided additional information to the median value and were not severely affected by imaging noise. The parameters derived by histogram analyses were insensitive to spatial heterogeneity and were strongly correlated to the median value, and the Haralick features were severely influenced by imaging noise and did not differentiate between untreated and sunitinib-treated tumors. The spatial gradient method was superior to histogram analyses and Haralick features for assessing intratumor heterogeneity in K trans maps of untreated and sunitinib-treated PDAC xenografts, and can possibly be used to assess intratumor heterogeneity in other medical images and to evaluate effects of other treatments as well.

Lane-Level Localization Using an AVM Camera for an Automated Driving Vehicle in Urban Environments

This paper describes a lane-level localization algorithm based on a map-matching method for application to automated driving in urban environments. The lane-level localization implies localizing the vehicle with centimeter-level accuracy. In order to achieve a satisfactory level of position accuracy with a low-cost GPS, a sensor fusion approach is essential for lane-level localization. The proposed sensor fusion approach for the lane-level localization of a vehicle uses an around view monitoring (AVM) module and vehicle sensors. The proposed algorithm consists of three parts: lane detection, position correction, and localization filter. In order to detect lanes, a commercialized AVM module is used. Since this module can acquire an image around the vehicle, it is possible to obtain accurate position information of the lanes. With this information, the vehicle position can be corrected by the iterative closest point (ICP) algorithm. This algorithm estimates the rigid transformation between the lane map and lanes obtained by AVM in real-time. The vehicle position corrected by this transformation is fused with the information of vehicle sensors based on an extended Kalman filter. For higher accuracy, the covariance of the ICP is estimated using Haralick's method. The performance of the proposed localization algorithm is verified via vehicle tests on a proving ground. Test results show that the proposed method can achieve localization centimeter-level accuracy. The proposed algorithm will be useful in the implementation of automated driving control.

MRI Sagittal Image Segmentation from Patients with Abdominal Aortic Aneurysms

Early detection in patients with abdominal aortic aneurysm (AAA) is esdential to reduce the risk of rupture of aortic wall that causes bleeding and often lead to death. Information about the condition of AAA is indispendable to complete the diagnosis of doctors in decision making. The position and shape of AAA can be obtained by sagittal image from an MRI examination. Characteristics of MRI sagittal image are having a gray level that is almost teh same between one organ to another. Therefore, to separate between one organ to another is difficult. This research is conducted MRI sagittal iamge segmentation in patients to obtain information on morphology and location of abdominal aortic aneurysm (AAA). To Segmenting the MRI Image we comobine thresholding method and Haralick Method. Under this proposed method, obtained sagittal images of the aorta are used to gain information about the location and shape of the aneurysm in abdominal aorta.

Parallel four‐dimensional Haralick texture analysis for disk‐resident image datasets

AbstractTexture analysis is one possible method of detecting features in biomedical images. During texture analysis, texture‐related information is found by examining local variations in image brightness. Four‐dimensional (4D) Haralick texture analysis is a method that extracts local variations along space and time dimensions and represents them as a collection of 14 statistical parameters. However, application of the 4D Haralick method on large time‐dependent image datasets is hindered by data retrieval, computation, and memory requirements. This paper describes a parallel implementation using a distributed component‐based framework of 4D Haralick texture analysis on PC clusters. The experimental performance results show that good performance can be achieved for this application via combined use of task‐ and data‐parallelism. In addition, we show that our 4D texture analysis implementation can be used to classify imaged tissues. Copyright © 2006 John Wiley & Sons, Ltd.

Is There Functional Vascular Information in Anatomical MR Sequences? A Preliminary in Vivo Study

According to the type of sequences used, either morphological or dynamic functional study can be performed using magnetic resonance imaging (MRI). The aim of this study is to find out if vascular information found, in dynamic MR sequences, already exists in anatomical MR sequences in the particular case of Legg-Calvé-Perthes disease (LCPD). LCPD is due to a loss of circulation to the femoral head in a growing child resulting in avascular necrosis and leading to possible distortion of size and shape of the proximal femur. MRI acquisitions consist in performing two anatomical sequences and one dynamic sequence with a gadolinium-diethylenetriamine pentaacetic acid (Gd-DTPA) injection. Five new parametric images characterizing hyper- and hypo-vascularized areas are computed from the dynamic MR sequence. For each new image, the two corresponding anatomical images are found and registered. Then, four types of regions of interest (ROIs) are extracted: healthy hyper- and hypo-vascularized areas and pathological hyper- and hypo-vascularized areas. First-order statistical parameters and texture parameters (Haralick's method, run length method, fractal parameters, autoregressive factors and Laws' texture energy method) are computed in each ROI. Then, a statistical study based on a T test is performed. Results show that some parameters could discriminate the four ROI types. Hence, dynamic vascular image and intrinsic anatomical image characteristics seem to be correlated. Finally, the disease can be evaluated with objective parameters using only anatomical sequences.

Comparison of texture analysis methods for the characterization of coronary plaques in intravascular ultrasound images

Intravascular Ultrasound (IVUS) is a diagnostic imaging technique that provides tomographic visualization of coronary arteries. The aim of this study was to evaluate five texture analysis techniques and determine their ability to distinguish between plaque lesions of different composition. Using histological correlation, regions of calcified, fibrous, and necrotic core plaque were chosen from 27 coronary plaques. First-order statistics, Haralick's method, Laws' texture energy method, the neighborhood gray-tone difference matrix method, and texture spectrum features were examined using discriminant analysis. Self-validation indicated that Haralick's method yielded the most accurate results, with resubstitution and cross-validation error rates of 0.00 and 14.76%, respectively. Further optimization gave error rates of 6.67%, using only two discriminating features, IDM and entropy.

Using Haralick method to measure wood surface roughness

Wood surface roughness is one of the important indexes of manufactured wood products. This paper presents an attempt to use Haralick method to evaluate manufactured wood surface roughness through the utilization of imaging processing and pattern recognition techniques. Through the use of the methodology presented in this paper, We can give features of wood surface by spatial dependence matrices. The preliminary results indicated that the method has the advantages of non-contact, three-dimensional, high-speed. This method can be used in classification and in-time measurement of manufactured wood products.

Adaptable edge quality metric

A new quality metric for evaluating edges detected by digital image processing algorithms is presented. The metric is a weighted sum of measures of edge continuity, smoothness, thinness, localization, detection, nd noisiness. Through a training process, we can design weights that optimize the metric for different users and applications. We have used the metric to compare the results of ten edge detectors when applied to edges degraded by varying degrees of blur and varying degrees and types of noise. As expected, the more optimum Laplacian-of-Gaussians (LoG) filter and Haralick's second derivative method outperform the simpler gradient detectors. At high SNR, Haralick's method is the best choice, although it exhibits a sudden drop in performance at lower SNRs. The LoG filter's performance degrades almost linearly with SNR and maintains a reasonably high level at lower SNRs. The same relative performances are observed as blur is varied. For most of the detectors tested, performance drops with increasing noise correlation. Noise correlated in the same direction as the edge is the most destructive of the noise types tested.

Comments on "Digital Step Edges from Zero Crossings of Second Directional Derivatives

In a recent paper,1 Haralick published an edge detection scheme that was supported, in part, by an evaluation against the Prewitt and the Marr-Hildreth (¿2G) operators. This evaluation led to the conclusion that Haralick's method performed the best, and the ¿2G operator performed the worst. The implementation of the ¿2G operator, on which this evaluation was based, differed significantly from that used by Marr and Hildreth. Evaluation of the performance of the Marr-Hildreth implementation of the ¿2G operator on similar images shows that this edge detection method in fact performs comparably to the Prewitt and Haralick operators.