Haralick texture features from apparent diffusion coefficient (ADC) MRI images depend on imaging and pre-processing parameters

Abstract

In recent years, texture analysis of medical images has become increasingly popular in studies investigating diagnosis, classification and treatment response assessment of cancerous disease. Despite numerous applications in oncology and medical imaging in general, there is no consensus regarding texture analysis workflow, or reporting of parameter settings crucial for replication of results. The aim of this study was to assess how sensitive Haralick texture features of apparent diffusion coefficient (ADC) MR images are to changes in five parameters related to image acquisition and pre-processing: noise, resolution, how the ADC map is constructed, the choice of quantization method, and the number of gray levels in the quantized image. We found that noise, resolution, choice of quantization method and the number of gray levels in the quantized images had a significant influence on most texture features, and that the effect size varied between different features. Different methods for constructing the ADC maps did not have an impact on any texture feature. Based on our results, we recommend using images with similar resolutions and noise levels, using one quantization method, and the same number of gray levels in all quantized images, to make meaningful comparisons of texture feature results between different subjects.

Quantitative variations in texture analysis features dependent on MRI scanning parameters: A phantom model

“…This work highlights the importance of using standardized and rigorously controlled scanning protocol when conducting research utilizing a texture analysis. This current study expands upon prior studies published in the literature which previously investigated a limited set of MRI acquisition parameters and their influence on texture features . The study performed by Mayerofer et al .…”

“…This current study expands upon prior studies published in the literature which previously investigated a limited set of MRI acquisition parameters and their influence on texture features . The study performed by Mayerofer et al . investigated changes in TR/TE, sampling bandwidth, and number of acquisitions and the influence of these parameters on texture analysis features.…”

“…investigated changes in TR/TE, sampling bandwidth, and number of acquisitions and the influence of these parameters on texture analysis features. Mayerofer et al., noted that changes in these features had a substantial impact on the sensitivity of the texture analysis features, however, this study examined a limited set of MRI scanning parameters which did not include a study of NEX, flip angle, magnet strength, and scanner platform (GE vs Siemens). This current study seeks to bridge the gap in knowledge investigating the influence these additional MRI scanning parameters have on certain texture analysis features.…”

“…Based on our knowledge of how changes in the CT scanning parameters varies texture analysis features, as well as preliminary studies investigating the sensitivity of texture features to variations in MRI technique, we could similarly deduce that changes in MRI scanning parameters such as differences in magnet strength and scanner platform could also influence texture analysis features. Thus, the purpose of this study was to evaluate and quantify changes in MRI sequence parameters may have on texture analysis features using a simple, nonanatomic phantom model.…”

Joint Prediction of Breast Cancer Histological Grade and Ki-67 Expression Level Based on DCE-MRI and DWI Radiomics

“…Although the gray-level quantization could affect the calculated texture features, different groups tend to use arbitrarily chosen quantization methods when constructing the GLCM [44]. A previous study analyzed quantization parameters of texture, and recommend the same number of gray levels in all quantized images for texture features [45].…”