A full convolutional neural network epilepsy focus segmentation method based on multi-mode images

Abstract

The invention discloses a full convolutional neural network epilepsy focus segmentation method based on a multi-mode image, and mainly solves the problem that the focus in the epilepsy image is difficult to segment in the prior art. The implementation method comprises the following steps: adjusting an original brain MRI image and a PET image to the same resolution space, carrying out edge cutting;dividing the clipped MRI / PET image into a training set and a test set; establishing full convolutional neural network Y-Net; icarrying out input of training sets to Y- Training in the Net network, and carrying out the trained Y-net network; storing the convolution kernel parameters of the convolution layer in the Net network; loading stored convolution kernel parameters to constructed Y-Net network, and inputting the test set to obtain an automatic segmentation result of the epileptic focus. The method has the advantages of easiness in segmentation and high segmentation precision, and can beused for segmenting epileptic focus areas in brain nuclear magnetic resonance imaging (MRI) and positron emission tomography (PET) images.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a multimodal image lesion segmentation method, which can be used to segment epileptic focus areas in brain magnetic resonance image MRI and positron emission tomography image PET. Background technique [0002] With the rapid development of computer technology and medical imaging technology, many medical imaging technologies have emerged, such as computed tomography CT, three-dimensional ultrasound imaging, positron emission tomography PET, magnetic resonance imaging MRI, single photon emission computed tomography SPECT, diffuse Weighted imaging DWI, functional magnetic resonance FMRI, etc. In clinical medicine, medical imaging plays an increasingly important role, especially magnetic resonance, brain CT and brain waves. Medical images have become an important tool and means for doctors to diagnose and treat patients. [0003] Since nuclear magnetic resonanc...

AI Technical Summary

Problems solved by technology

The edge-based level set model mainly uses edge information to segment images. This model is sensitive to initialization conditions and noise. When the target edge is blurred, its segmentation effect is not satisfactory.

Although the region-based level set segmentation model is not sensitive to noise, it is easy to over-segment

The segmentation methods mentioned above often require a lot of prior knowledge, and for weak targets such as epileptic lesions, it is difficult to provide effective prior information, and the segmentation results are prone to over-segmentation, and it is easy to divide healthy brain tissue into lesion areas. Induce doctors to damage healthy brain tissue during surgery, which not only reduces the therapeutic effect of epilepsy surgery, but also destroys the physiological function of healthy brain tissue, resulting in severe surgical complications

Images