ABSTRACT

Hypertension, or increased blood pressure (BP), may injure blood veins in the retina of an eye, resulting in a condition known as hypertensive retinopathy (HR). Hypertension causes blood to enlarge as well as the retina's function to deteriorate. The most common way to identify HR in a person's blood is by a medical assessment using an ophthalmoscope that is still done manually by an ophthalmologist. It takes a very some duration for a physician to detect a particular member on the eye fundus picture in a very manual method. To solve this challenge, a mechanism for automatically identifying the picture of the retinal fundus is required. The backpropagation bp neural network was utilized in this study to identify the retinal fundus. Preprocessing green channel, contrasting limit adaptive entropy equalization, morphology proximity, background subtraction, threshold, or linked components analysis, and extracted features utilizing zoned were done before recognition. The findings demonstrate that perhaps the suggested approach can immediately recognize the retinal fundus 95% of such time with a maximal period of 1500. Keywords: Machine Learning, Hypertension detection, Classification, Blood pressure, Green channel; Healthcare
Publication date: 01/11/2021

https://ijbpas.com/pdf/2021/November/MS_IJBPAS_2021_NOV_SPCL1078.pdf

Download PDF
https://doi.org/10.31032/IJBPAS/2021/10.11.1078