A recursive support vector machine (RSVM) algorithm to detect and classify diabetic retinopathy in fundus retina images

Kanagasabai, Malathi; Nedunchelian, R.

doi:10.4066/biomedicalresearch.29-16-2328

Cited by 19 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some of the previous studies by (Prabakaran and Kannadasan 2018), who states that deep learning technology has more accuracy of about 80% due to hidden layer classification among specific medical parameters out of 64 but the results were identified in the proposed model based on the contrast with Decision tree algorithm, which has the highest accuracy of 98% due to top-down classification during preprocessing. The image processing is involved in the detection of images on a heart disease predict relevant details based on the detection by different classification process (("Recognition and Classification of Diabetic Retinopathy Utilizing Digital Fundus Image with Hybrid Algorithms" 2019; Malathi and Nedunchelian 2018). The findings in this paper were almost similar to the above-cited papers.…”

Section: Discussionsupporting

confidence: 75%

An Innovative Method for Predicting and Classifying Inadequate Accuracy in Heart Disease by Using Decision Tree with K-Nearest Neighbors Algorithm

Rajesh¹,

Malathi²

2021

alinteri

View full text Add to dashboard Cite

Aim: Predicting the Heartdiseases using medical parameters of cardiac patients to get a good accuracy rate using machine learning methods like innovative Decision Tree (DT) algorithm. Materials and Methods: Supervised Machine learning Techniques with innovative Decision Tree (N = 20) and K Nearest Neighbour (KNN) (N = 20) are performed with five different datasets at each time to record five samples. Results: The Decision Tree is used to predict heart disease with the help of various medical conditions, the accuracy is achieved for DT is 98% and KNN is 72.2%. The two algorithms Decision Tree and KNN are statistically insignificant (=.737) with the independent sample T-Test value (p<0.005) with a confidence level of 95%. Conclusion: Prediction and classification of heart disease significantly seem to be better in DT than KNN.

show abstract

Section: Discussionsupporting

confidence: 75%

An Innovative Method for Predicting and Classifying Inadequate Accuracy in Heart Disease by Using Decision Tree with K-Nearest Neighbors Algorithm

Rajesh¹,

Malathi²

2021

alinteri

View full text Add to dashboard Cite

show abstract

“…In this report, they used the unsupervised learning method of the K-means algorithm for pattern recognition of the iris flower dataset [31]. [32,33]. In this article, the machine itself recognises the species of iris flower by applying unsupervised learning to neural network algorithms [34] This paper observed that a multilayer feed-forward neural network gives high and good accuracy results in the iris flower data set [35].…”

Section: Discussionmentioning

confidence: 99%

Decision Tree Over Support Vector Machine for Better Accuracy in Identifying the Problem Based on the Iris Flower

Poojitha

Kanagasabai

2022

Advances in Parallel Computing Algorithms, Tools and Paradigms

View full text Add to dashboard Cite

The iris dataset will be classified using the support vector machine and decision tree algorithms. flower dataset identifies the pattern and classifies it. The dataset has 150 rows and 5 attributes, which contains 50 samples from each species. There are three species in this dataset. Iris flower classification can be performed using support vector machines and decision tree algorithms. SVM stands for Support Vector Machine, and is a supervised machine learning technique that can be used for classification and regression. The Decision Tree algorithm is a simple approach mainly used for classification and prediction. The sample size has been determined to be 20 for both the groups using G Power 80%. The Support Vector Machine algorithm provides a mean accuracy of 98.09% when compared to the Decision Tree algorithm, with a mean accuracy of 95.55%. A statistically insignificant difference was observed between the Decision Tree and the Support Vector Machine, p = 0.92 (> 0.05) based on 2-tailed analysis. In the classification of Iris flowers, the Support Vector Machine outperformed the Decision Tree Algorithm.

show abstract

“…• The behavioral difference between an expert and no expert groups is investigated. • An automated labeling system to distinguish target and non-target regions from eye gaze behavior of expert optometrists and nonexpert groups while viewing fundus retinal images has been developed [19]. • Proposed a novel target detection system that combines bottom-up (BU) and top-down (TD) approaches.…”

Section: Contributions Of the Proposed Systemmentioning

confidence: 99%

Automated Detection of fundus retinal image using EGODD algorithm

Subha¹,

Mohideen²

2022

EAI Endorsed Trans Perv Health Tech

View full text Add to dashboard Cite

The developed device is tested for optical disc identification in the retinal image of the fundus. The theme chosen was interesting, as optical disc detection plays a critical role in retinal image processing for recognizing other fundus structures and is essential for identifying eye-related diseases. Because human visual vision has been little researched for optical disc detection and an effort to extract data analytics based on eye eyes, the proposed study first discovers how human perception functions for optical disc detection using a bottom-up visual focus paradigm. Eye-view data as the user conducts basic target-search tasks are obtained from separate user groups made up of experts and no expert groups. Extensive data processing has been carried out to extract eye gazing characteristics such as fixation and using computer approach mark regions in the fundus' retinal picture. Segregated labeled data was used to build up a piece of top-down information to bend the search map to the target area. The resulting Eye-based Optical Disk Detection System (EGODD) was tested through normal machine learning algorithms. This proposed framework has been testedand approved for optical disc detection; however, it can also be applied to other applications.

show abstract

A recursive support vector machine (RSVM) algorithm to detect and classify diabetic retinopathy in fundus retina images

Cited by 19 publications

References 14 publications

An Innovative Method for Predicting and Classifying Inadequate Accuracy in Heart Disease by Using Decision Tree with K-Nearest Neighbors Algorithm

An Innovative Method for Predicting and Classifying Inadequate Accuracy in Heart Disease by Using Decision Tree with K-Nearest Neighbors Algorithm

Decision Tree Over Support Vector Machine for Better Accuracy in Identifying the Problem Based on the Iris Flower

Automated Detection of fundus retinal image using EGODD algorithm

Contact Info

Product

Resources

About