Magnetic Resonance Imaging Manifestations of Pediatric Purulent Meningitis Based on Immune Clustering Algorithm

Abstract

The purpose of this study was to analyze the diagnostic value of magnetic resonance imaging (MRI) based on the immune clustering algorithm (ICA) in children with purulent meningitis. In this study, 235 children with suspected pediatric purulent meningitis (PPM) were routinely scanned, and the artificial immune algorithm (AIA) and ICA were applied to image processing. In order to quantitatively analyze the accuracy and precision of the processed image, precision rate was introduced as the evaluation of accuracy, and the True Positive Vis Fox, False Negative Vis Fo, and False Positive Vis Fo were selected as the evaluation indicators. After comparison, the accuracy, sensitivity, and specificity of ICA detection were higher than those of AIA and conventional plain scanning, and the differences were statistically obvious ( $P<0.05$ ). Comparison on image display effects showed that compared with AIA, the image processed by the ICA algorithm constructed in this study showed the highest definition and contrast and the best denoising effect and image quality, showing a statistically obvious difference ( $P<0.05$ ). All in all, the display effect of MRI images of pediatric purulent meningitis based on ICA was more accurate and clearer than that of the traditional image processing, and it can provide a more accurate auxiliary basis in the diagnosis of lesion details. It also showed a higher clinical value for the development of a diagnosis and treatment plan for complicated PPM.