Measuring method of size of intracranial hematoma

Abstract

The invention provides a measuring method of the size of intracranial hematoma. As for characteristics of images of intracranial hematoma, non-linear anisotropic diffusion filtering is adopted for pre-processing of the images, noise and image artifacts are effectively removed, and actual requirements for medical image processing are met; two-step image segmentation processing is adopted for the images, non-brain tissue of the skull and outside the skull is removed ingeniously, and image registering and other work are not needed; the concept of entropy is introduced in image segmentation, two-dimensional thresholding of entropy is achieved, the images are segmented by using the gray level of pixels and the neighborhood gray level of pixels as parameters, information distribution of the gray level of the pixels is reflected, and relative information of neighborhood space of the pixels of the images is reflected. Therefore, noise is effectively restrained, and interference and undesired fine structures are removed; various operators of the genetic algorithm are improved, a self-adaption large-variation operator is adopted for solving the problem that the genetic algorithm has precocity, and namely the algorithm is constricted to a non-overall optimal point untimely, and the best threshold value is searched for through the improved layering genetic algorithm finally.

Description

technical field [0001] The research on the intracranial hematoma volume measurement method of the present invention is to convert the clinical requirement of intracranial hematoma volume measurement into the image hematoma segmentation problem, and propose a new segmentation algorithm, which belongs to the field of clinical application software. Background technique [0002] Brain injury leads to intracranial hemorrhage, and when blood accumulates in the brain to a certain volume, an intracranial hematoma is formed. Usually, a supratenorial hematoma of more than 20ml and an infratentorial blood volume of more than 10ml can cause symptoms of acute craniocerebral compression, which can be life-threatening. Hypertensive cerebral hemorrhage of more than 30ml is a disease with a high morbidity and disability rate in middle-aged and elderly people. Generally, the hematoma on the tentorium is more than 30ml, and the hematoma on the infratentorial is more than 10ml. Combined with ot...

AI Technical Summary

Problems solved by technology

However, due to the complexity of medical images and the differences between different individuals, at present, no segmentation algorithm can produce satisfactory segmentation results for all images, and at the same time, there is no specific type of image that can make all algorithms can obtain equally good segmentation results