Retina pathological image analysis system based on machine learning

Abstract

The invention provides a retina pathological image analysis system based on machine learning. The system comprises a retina original image sample acquisition module, an image graying processing module, a histogram specification module, an image normalization module, a nerve fiber layer marking module, a cell nucleus marking module, an image cutting module, a nerve fiber layer network training module, a cell nucleus network training module, a nerve fiber layer prediction module and a cell nucleus prediction module. The retina pathology image analysis system based on machine learning has the advantages that the retina pathology image analysis system based on machine learning has the following advantages. The retinal pathological image analysis system based on machine learning provided by theinvention is small in artificial participation amount, is suitable for large sample size analysis, has the advantages of high recognition speed and high recognition accuracy, and has more effective quantitative analysis and practical application prospects.

Description

technical field [0001] The invention belongs to the technical field of combination of biomedicine and computer, and in particular relates to a retinal pathology image analysis system based on machine learning. Background technique [0002] There are many causes of retinopathy, including retinal vascular disease, retinal congenital abnormalities, retinal circulation disorders, and retinal vasculitis. Clinically, there are many types of retinopathy, including abnormal retinal vascular disease, diabetic retinopathy, arteriosclerosis and hypertensive retinopathy, retinal vascular disease caused by systemic diseases, retinal hemangioma, traumatic vascular retinopathy, poisoning And radiation retinopathy and choroidal disease, hemangioma, and various degenerative diseases. [0003] In the prior art, after OCT images the retina, the retinal parameters related to retinopathy can only be observed by the doctor with the naked eye, which relies heavily on the doctor's personal experie...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, after OCT images the retina, the retinal parameters related to retinopathy can only be observed by the doctor with the naked eye, which relies heavily on the doctor's personal experience, and there is no uniform, objective and accurate test result, which affects ophthalmology clinical practice. diagnostic accuracy

Similarly, when conducting pathological research on animal retinas, the existing technology is to prepare animal retinas as HE-stained sections, and judge pathological changes through manual interpretation by pathologists. This is highly subjective, and different pathologists have different reading results. , and cannot be quantitatively analyzed

[0004] The name of the patent applied by the inventor is an invention patent for a quantitative analysis system for retinal lesions based on image recognition. Although the effect of quantitative analysis on retinal images can be achieved, the analysis system is based on a small sample size quantitative analysis system. The analysis speed of unknown retinal images is low, which is not suitable for pathological analysis of a large number of retinal images

Images