Image Texture Characterization Using the Discrete Orthonormal S-Transform

Abstract

We present a new efficient approach for characterizing image texture based on a recently published discrete, orthonormal space-frequency transform known as the DOST. We develop a frequency-domain implementation of the DOST in two dimensions for the case of dyadic frequency sampling. Then, we describe a rapid and efficient approach to obtain local spatial frequency information for an image and show that this information can be used to characterize the horizontal and vertical frequency patterns in synthetic images. Finally, we demonstrate that DOST components can be combined to obtain a rotationally invariant set of texture features that can accurately classify a series of texture patterns. The DOST provides the computational efficiency and multi-scale information of wavelet transforms, while providing texture features in terms of Fourier frequencies. It outperforms leading wavelet-based texture analysis methods.

Predicting MGMT Methylation Status of Glioblastomas from MRI Texture

“…al. [5], indicates that the rotationally invariant DOST outperforms leading waveletbased texture analysis methods. The spatial-frequency technique extracts texture features by decomposing an MR image into a set of images at various spatial frequencies.…”

“…(5) Based on the 2D-DOST, a rotationally invariant spectrum, was created by averaging specific frequency orders together [5]. The process is depicted in Figure 2.…”

“…Motivated by these findings, this research presents an automated method for predicting the methylation status of the GBM based on texture analysis of T2, FLAIR and T1-postcontrast MRI scans. In order to analyze the MR images, the proposed system utilizes the 2-dimensional discrete orthogonal S-transform [5] to extract texture features that are subsequently used by an 1 -Regularized neural network to predict the methylation status of a given GBM. In a leave-one-out cross validation study, the proposed system achieved an average accuracy of 87.7%, high enough for use in clinical diagnosis.…”

Pilot Study: Evaluation of Dual-Energy Computed Tomography Measurement Strategies for Positron Emission Tomography Correlation in Pancreatic Adenocarcinoma

“…At the time of this study, no automated or semiautomated software programs were available to measure pancreatic tumors during DECT to decrease measurement variability. Additionally, SD and SUV max are relatively crude measures of heterogeneity; more sophisticated measures of tumor texture, such as those based on the Stockwell transform [22,23] could be used as well. A more objective measurement approach such as texture analysis may ultimately achieve more reliable and reproducible results than our subjective placement of ROIs; however, these techniques are not currently commercially available.…”

MRI Texture Analysis Predicts p53 Status in Head and Neck Squamous Cell Carcinoma

“…25 Studies in glioblastoma have shown that there is a correlation between the methylation of O6-methylguanine-DNA methyltransferase and MR imaging features. 26 Levner et al 27 extracted texture features from MR images by using spatial frequency analysis and the Stockwell transform (ST) representation 28 and fed these characteristics into a neural network to predict the methylation status with an average accuracy of 87.7%. 27 …”