Susceptibility of texture measures to noise: an application to lung tumor CT images

TL;DR

This study evaluates how five different texture analysis methods respond to noise in lung tumor CT images, highlighting the robustness of autocovariance and the sensitivity of GLCM to noise.

Contribution

It systematically compares the susceptibility of five texture measures to noise in lung tumor CT images, providing insights into their robustness.

Findings

Autocovariance measure is least affected by noise.

Gray level co-occurrence matrix is most affected by noise.

ROI size influences the number of features and susceptibility to noise.

Abstract

Five different texture methods are used to investigate their susceptibility to subtle noise occurring in lung tumor Computed Tomography (CT) images caused by acquisition and reconstruction deficiencies. Noise of Gaussian and Rayleigh distributions with varying mean and variance was encountered in the analyzed CT images. Fisher and Bhattacharyya distance measures were used to differentiate between an original extracted lung tumor region of interest (ROI) with a filtered and noisy reconstructed versions. Through examining the texture characteristics of the lung tumor areas by five different texture measures, it was determined that the autocovariance measure was least affected and the gray level co-occurrence matrix was the most affected by noise. Depending on the selected ROI size, it was concluded that the number of extracted features from each texture measure increases susceptibility to noise.