Diagnostic efficacy of computed tomography histogram analysis for the differentiation of histopathological low- and high-grade tumors in colocteral carcinoma

Abstract

Aim: Colorectal adenocarcinoma (CA) is the most common type of cancer worldwide and the third leading cause of cancer-related deaths. Primary pathological grade bears importance in the course of the disease. The possibility of non-invasive grading through radiology modalities is still an important issue. The present study aims to reveal whether a non-invasive grading similar to pathological grading can be performed using histogram analysis on computed tomography (CT) scan images. Material and Method: 58 patients operated and diagnosed with CA pathologically were included in the present study. As for medical protocol, abdominal intravenous contrast CT scan images obtained from TOSHIBA Alexion and TOSHIBA Aquilion ONE (Toshiba Medical Systems, Nasu, Japan) devices with 120 kVp tube voltage were set to a window width of 400 and a window level of 40. Patient images from retrospective scanning were evaluated on a workstation. For the evalution of mass, intraluminal air, necrotic areas, pericolonic fat tissue or intra-mass large feeding vessels were not included in the measurement range. Mass size was measured on the largest axis according to the longest axis. For histogram analysis, regions of interest were positioned. Parameters included in the histogram analysis were pixels, mean, standard deviation, minimum, maximum, median, variance, entropy, size L%, size U%, size M%, kurtosis, skewness, uniformity, percent01, percent03, percent05, percent10, percent25, percent75, percent90, percent95, percent97 and percentile 99. Results: Histogram analysis results obtained from three different measurements for each of 58 patients were not found to be statistically significant in the differentation of pathologically defined histological grading system. Conclusion: Although the use of a non-invasive method instead of an invasive one may offer an advantage, was not statistically significant in the prediction of histological grade.