A Survey of Methods for Genome Functional Analysis in Comparative Genomics

Diabetic retinopathy is one of the major eye problems caused by defects in the blood vessels that occur in retinal tissues. Uncontrolled blood sugar makes diabetic retinopathy. Majority of the diabetic patients are affected by Macula Edema problems which makes vision loss. Diabetic Macula Edema can be identified at its earlier stage by floaters, dark areas, and blur in the vision. Earlier stages of the DME can be treated and saved from vision loss. The flow or deposit of the blood make swelling in the macula and it makes a serious complication called DME. Ophthalmologists used various computer methods to diagnose fundus images to identify macular Edema and its stages, but the accuracy of the DME detection and classification using fundus images with traditional methods is less. This problem is considered as a major problem and most medical and healthcare experts are trying to provide a better solution with high accuracy. This paper has aimed to develop a combination of the U-Net model with the Multi-Class Support Vector Machine for segmenting and classifying OCT images. The experimental results show that the proposed U-Net-MSVM provides high accuracy in terms of classification. The performance of the proposed model is evaluated by comparing its result with earlier methods.

show abstract

A Survey of Methods for Genome Functional Analysis in Comparative Genomics

Cited by 3 publications

References 7 publications

Machine Learning Approach to Lung Cancer Survivability Analysis

Machine Learning Approach to Lung Cancer Survivability Analysis

A combined U-Net and multi-class support vector machine learning models for diabetic retinopathy macula edema segmentation and classification DME

A Combined U-Net and Multi-Class Support Vector Machine Learning Models for Diabetic Retinopathy Macula Edema Segmentation and Classification DME

Contact Info

Product

Resources

About