Glaucoma optic disc segmentation map obtaining method

Abstract

The invention discloses a glaucoma optic disc segmentation map obtaining method. The method comprises the steps of acquiring and classifying a known color fundus image and a fundus image data set; constructing an optic disc preliminary segmentation deep learning network, and training, testing and correcting the network by utilizing the classification data to obtain an optic disc segmentation model; carrying out segmentation and screenshot on the test data set and the training data set; processing the screenshot; constructing an interpretable glaucoma preliminary auxiliary network; utilizing the screenshot to train and correct the interpreted glaucoma preliminary auxiliary network to obtain an interpretable glaucoma auxiliary network; segmenting the to-be-analyzed color fundus image data byusing an optic disc segmentation model to obtain an optic disc screenshot; processing the view disk screenshot by using an interpretable glaucoma auxiliary network to obtain a gray scale thermodynamic diagram; And processing the gray scale thermodynamic diagram to obtain a final view disk segmentation map. According to the method, the optic disc segmentation map for auxiliary diagnosis can be rapidly provided for doctors, and the method is high in reliability and good in effect.

Description

technical field [0001] The invention specifically relates to a method for obtaining an optic disc segmentation map of glaucoma. Background technique [0002] With the development of society, economy and technology, people pay more and more attention to health. And along with the popularity of smart electronic devices (such as smart phones, tablet computers, etc.) and the popularity of large-screen smart electronic devices, smart electronic devices have gradually occupied most of people's entertainment and leisure time. [0003] With the popularity of smart electronic devices, especially large-screen smart electronic devices, the incidence of glaucoma is increasing year by year. Glaucoma is the second leading cause of blindness in the world. The blinding symptoms it causes are difficult to detect. The symptoms are generally blindness from the outside to the inside. When the patient is aware of it, it is already in the late stage. At present, glaucoma cannot be cured, and t...

AI Technical Summary

Problems solved by technology

[0006] However, the current auxiliary screening method is not widely used, the reason is that the current method can only give the results of auxiliary screening, and once the model is wrong or the performance of the model is not good, the reliability of the model will be reduced. Not high; and although the current method can give the corresponding auxiliary screening results, it cannot give the corresponding auxiliary diagnosis basis and interpretability information, making the current method not high in credibility

Images