The Hybrid Method of SOM Artificial Neural Network and Median Thresholding for Segmentation of Blood Vessels in the Retina Image Fundus

Abstract

Blood vessels in the retina of the eye are one important sign when making a diagnosis of hypertensive retinopathy. On the retina can be known several signs including tortuosity and arteriovenous ratio. Blood vessels mixed with a number of objects in the retina, the segmentation of blood vessels becomes a very interesting challenge because they have to separate blood vessels from a number of objects. This study aims to segmentation blood vessels using the main method of self-organizing maps artificial neural networks (SOM-ANN). The proposed segmentation method is divided into three stages, namely preprocessing, segmentation, and performance analysis. The preprocessing step is to improve image quality using the contrast-limited adaptive histogram equalization (CLAHE), median filter, and morphology. The segmentation stage uses the SOM-ANN algorithm combined with the mean or median thresholding. The performance parameters which are measured consist of sensitivity, specificity, and area under the curve (AUC). The test results using the dataset STARE and DRIVE show that the median thresholding is able to provide the best AUC performance compared to the mean thresholding. The proposed segmentation model is able to provide performance in the excellent category, with AUC values of 90.55% for the STARE dataset and 90.35% for the DRIVE.

[Retracted] Analysis of Enterprise Human Resources Demand Forecast Model Based on SOM Neural Network

“…e development of unsupervised learning is still immature, and the SOM algorithm still has some limitations [19][20][21][22] as follows:…”

“…The development of unsupervised learning is still immature, and the SOM algorithm still has some limitations [ 19 – 22 ] as follows: The network structure is fixed and cannot be changed dynamically During network training, some neurons can never win and become “dead neurons” The SOM network cannot add new categories before complete relearning When the input data is small, the training result usually depends on the sample The initial state of the network connection right and the parameter selection in the algorithm have a greater impact on the convergence performance of the network …”

An event‐driven fusion framework with auto‐scaling of edge intelligence for resilient smart applications in developing countries

“…Data coming with higher arrival rates can be down‐scaled using median as a threshold. Assuming there are $$$ n $$$ values in array $$$ microservice\_ array $$$ where each value is the number of containers required with every data, the $$$ n $$$ values are sorted and the median is calculated as: 37 $$$$ \mathrm{median}\in \left\{\begin{array}{cc} microservice\_ array\left[\frac{n+1}{2}\right]\kern1em \hfill & \mathrm{when}\kern0.3em n\kern0.3em \mathrm{is}\ \mathrm{odd}\hfill \\ {}\frac{microservice\_ array\left[\frac{n}{2}\right]+ microservice\_ array\left[\frac{n+1}{2}\right]}{2}\kern1em \hfill & \mathrm{when}\kern0.3em n\kern0.3em \mathrm{is}\ \mathrm{even},\hfill \end{array}\right. $$$$ Maximum likelihood estimation: This technique is applied when there are sufficient number of samples with known probability distribution as normal.…”

The Relationship between Social Entrepreneurship Capability of SOM Neural Network Algorithm and New Enterprise Performance

“…Suryani and Susilo aim to segment blood vessels using the main method of self-organizing map artificial neural networks. e segmentation method they proposed can effectively improve the test performance of medical machinery and increase the success rate of surgery [3]. Runst et al proposed a novel classification scheme aimed at improving the identification of entrepreneurial companies in the microcensus.…”