Application Of DICOM Format CT Image Segmentation Technology In Cerebral Hemorrhage Imaging Examination

In recent years,medical imaging has developed rapidly.With the birth of the DICOM standard,the advancement of medical automation and informatization has been promoted.In order to further optimize medical images and improve readability,clinicians can better grasp the lesions and improve the accuracy of diagnosis.Rate,people began to pay attention to the processing of medical images,and formed an important subject.Taking cranial CT as an example,this study is based on the application of image segmentation technology in DICOM format,which provides clinical support for cerebral hemorrhage and cranial CT.Based on the FCM clustering algorithm and the Laplacian of Gaussian,this paper realizes the analysis of CT images of the brain hemorrhage.Image segmentation adopts two-level segmentation of coarse segmentation processing(edge and intracranial structure extraction)and subdivision processing(image segmentation and hematoma measurement)to ensure the quality of image segmentation and improve the recognition effect of cerebral hemorrhage and brain CT imaging.The research focuses on the application of the LOG operator,and further optimizes the FCM clustering fuzzy algorithm.In order to further confirm the performance advantages of the improved FCM fuzzy clustering algorithm,this section adopts a quantitative analysis method to select 45 patients(96 lesions)with cerebral hemorrhage for the test,and divide them into three groups according to the number of lesions.Group A uses the standard FCM algorithm A total of 13 cases(32lesions)were segmented;group B was segmented using FCM＿S algorithm,a total of 14cases(32 lesions);group C was segmented using improved FCM algorithm,a total of 18cases(32 lesions).The optimized algorithm is analyzed through experiments.From the comparison results,it can be seen that when the improved FCM algorithm processes Gaussian noise images,the running time and the number of clustering are reduced,respectively(8.87±2.35)s and(30.12±3.51)times When processing salt and pepper noise images,the running time of the improved FCM algorithm,the number of clustering(8.57±2.39)s,(32.67±3.46)times are close to the standard FCM algorithm,and they are all lower than the FCM＿S algorithm;it can be seen that the improved algorithm After optimization,the number of clustering is reduced,the time-consuming problem of standard FCM algorithm and FCM＿S algorithm is reduced,and it can maintain good anti-noise performance and high stability.The improved FCM algorithm can obtain the complete and void-free part of the brain parenchyma by segmenting the image,ensuring the smooth boundary and avoiding the influence of the special structure of the brain edge.After comparison,the accuracy and sensitivity of image segmentation under the improved FCM algorithm were 0.976 and 0.734,which were significantly higher than those of the other two groups.The experimental results confirmed the improved effect of the FCM algorithm and provided a theoretical basis for future related research.Through the comparative analysis of experiments,the conclusions are as follows: 1.Intracerebral hemorrhage CT applies the edge segmentation method to improve the LOG operator,enhance the effect of edge extraction,and lay the foundation for further image segmentation processing.2.The CT image of cerebral hemorrhage is segmented by the improved FCM algorithm,which can evaluate the area and volume of the hematoma and provide a reference for clinical treatment.