Use of ADC histogram in identification of glioblastoma and solitary metastatic brain tumor

Abstract

Objective To investigate the value of using the ADC histogram in the identification of glioblastoma and solitary metastatic brain tumor. Methods Thirty cases of glioblastoma and 28 cases of solitary metastatic brain tumor, pathologically confirmed at the First Affiliated Hospital of Zhengzhou University between May 2013 and October 2017, were retrospectively analyzed. The MaZda software was used to extract the region of interest (ROI) from the tumor. Nine parameters of histogram were obtained and statistically analyzed. The ROC curve was generated to evaluate the value of the histogram parameters in the identification of glioblastoma and solitary metastatic brain tumor. Results Among the characteristics of the 9 parameters extracted from histogram, the differences in the mean, skewness, the 10th, 50th, and 90th percentiles between the two groups were statistically significant (P<0.05) . The areas under the ROC curve were 0.742, 0.679, 0.678, 0.751, and 0.709, respectively. According to the identification for the two tumors by the mean, the sensitivity and the specificity were 73.3% and 72.4%, respectively; the skewness was 60.7% and 60.0%; the 10th percentile was 70.0% and 60.7%; the 50th percentile were 63.3% and 78.6%; and the 90th percentile was 76.7% and 67.9%, respectively. Conclusion The ADC histogram analysis can provide more characteristics of quantitative information, and therefore represents a new method for the identification of glioblastoma and solitary metastatic brain tumor. Key words: Glioblastoma; Metastatic brain tumor; Histogram; Magnetic resonance imaging; Apparent diffusion coefficient