Hematoma Segmentation And Measurement System Based On CT Image

Intracerebral hemorrhage is a brain injury disease that seriously endangers human life and health.In the display of intracerebral hematoma,CT image is better than other images,so it is often used in the diagnosis of cerebral hemorrhage.Currently in clinical practice,doctors measure the amount of intracerebral hemorrhage of patients by manually sketching the outline of hematoma in CT images,so as to determine the severity of the disease.This method is not only cumbersome,but also very dependent on the clinical experience of doctors,so the measurement results may not be objective.In order to solve this problem,according to the characteristics of CT images of intracerebral hemorrhage,this paper studies the segmentation algorithm of cerebral hematoma,and designs and completes the manual interactive software of intracerebral hemorrhage measurement based on the segmentation algorithm,so as to realize the accurate segmentation and measurement of intracerebral hematoma.The details contents are as follows:（1）Based on the kernel intuitionistic fuzzy c-means clustering（KIFCM）algorithm,taking the neighborhood information of the image as the correction term of the objective function,an spatial kernel intuitionistic fuzzy c-means（SKIFCM）algorithm combined with neighborhood information is proposed to remove the small non-hematoma area which is close to the gray value of hematoma in intracerebral hemorrhage CT image;Combined with the location information of cerebral hematoma,the location weight factor is introduced in the clustering process,and a spatial kernel intuitionistic fuzzy c-means based on Gaussian Model（SKIFCM＿G）algorithm is proposed to improve the accuracy of cerebral hematoma segmentation;The membership correction method is used to optimize the clustering results,and a spatial kernel intuitionistic fuzzy c-means based on membership correction（SKIFCM＿M）algorithm is proposed to further improve the segmentation accuracy of cerebral hematoma.By comparing the segmentation results of the algorithms proposed in this paper with the gold standard given by doctors,the effectiveness of the algorithms is verified.（2）In order to further combine the research work with clinical practice,this paper designs and completes the intracerebral hemorrhage measurement software system.The software can realize the common image processing function of CT slice sequence of patients with cerebral hemorrhage,and realize the segmentation of cerebral hematoma and the measurement of cerebral hemorrhage volume by selecting points with the mouse,thus providing a new choice for the current clinical measurement of cerebral hemorrhage.