scispace - formally typeset
Search or ask a question
Author

V. Seenivasagam

Bio: V. Seenivasagam is an academic researcher from National Engineering College. The author has contributed to research in topics: Image segmentation & Grid. The author has an hindex of 8, co-authored 26 publications receiving 219 citations.

Papers
More filters
Journal ArticleDOI
01 Jul 2013
TL;DR: In this article, the authors summarized the commonly used techniques for heart disease prediction and their complexities are summarized in this paper and observed that Hybrid Intelligent Algorithm improves the accuracy of the prediction system.
Abstract: The Healthcare industry generally clinical diagnosis is done mostly by doctor's expertise and experience. Computer Aided Decision Support System plays a major role in medical field. With the growing research on heart disease predicting system, it has become important to categories the research outcomes and provides readers with an overview of the existing heart disease prediction techniques in each category. Neural Networks are one of many data mining analytical tools that can be utilized to make predictions for medical data. From the study it is observed that Hybrid Intelligent Algorithm improves the accuracy of the heart disease prediction system. The commonly used techniques for Heart Disease Prediction and their complexities are summarized in this paper.

71 citations

Journal ArticleDOI
TL;DR: Outstanding outcomes reveal that extracting properties of features extracted from the lung cancer images successfully and SVM combined with binary classification, even classification works better with Multi-class rather than SVM, therefore, may be considered as a promising tool to diagnose the stages of nodules and classify the severity of cancer.
Abstract: With the fast pace in collating big data healthcare framework and accurate prediction in detection of lung cancer at early stages, machine learning gives the best of both worlds. In this paper, a streamlining of machine learning algorithms together with apache spark designs an architecture for effective classification of images and stages of lung cancer to the greatest extent. We experiment on a combination of binary classification (SVM-non linear SVM with Radial Basis Function RBF) and Multi-class classification (WTA-SVM winner-takes-all with support vector machine) with threshold technique (T-BMSVM) to classify nodules into malignant or benign nodules and also their malignancy levels respectively. The dataset used for processing is sputum cell images that have been collected from microscope lab images. We have argued for handling and processing large sizes of data sets as sputum cell images in the field of classification using the map-reduce framework in MATLAB and Pyspark, which works better with Apache spark. Our approach outperforms the other methods by achieving stability even in increasing dataset size in leaps and bounds and with a minimum error rate. It achieves 86% accuracy and other metrics are AUC-0.88, misclassification rate through which it was proved that Support Vector Machine (SVM) outperforms other classifiers. These outsourced outcomes reveal that extracting properties of features extracted from the lung cancer images successfully and SVM combined with binary classification, even classification works better with Multi-class rather than SVM, therefore, may be considered as a promising tool to diagnose the stages of nodules and classify the severity of cancer. Also, Scalability and convergence analysis embed to prove the improving results of multi-class classification than SVM.

35 citations

Journal ArticleDOI
01 Dec 2020
TL;DR: An automatic CAD system is presented in three stage, automatic liver segmentation and lesion’s detection is carried out, and liver lesions classification into malignant and benign is done by using the novel contrast based feature-difference method.
Abstract: The liver is essential for endurance and to carry out a large number of significant functions, including manufacture of indispensable proteins, and metabolism of fats and carbohydrates. The examination of CT might be employed for planning and managing the treatments for tumor in a proper way and for directing biopsies as well as other simply determined process. The Manual segmentation and Computed Axial Tomography (CT) image classification is a tedious task and time consuming process for large amount of data. Computer-Aided Diagnosis (CAD) systems take part in a fundamental role in the detection of liver disease in an early stage and therefore decrease death rate of liver cancer. In this paper an automatic CAD system is presented in three stage. In the first step, automatic liver segmentation and lesion’s detection is carried out. Then, the next step is to extract features. At last, liver lesions classification into malignant and benign is done by using the novel contrast based feature-difference method. The extracted features from the lesion area with its surrounding normal liver tissue are based on intensity and texture. The lesion descriptor is obtained by considering the difference between the features of both lesion area and normal tissue of liver. Finally to categorize the liver lesions into malignant or benign a new SVM based machine learning classifier is trained on the new descriptors. The investigational outcome show hopeful improvement. Besides, the projected approach is insensitive to ranges of textures and intensity between demographics, imaging devices, and patients and settings. The classifier discriminates the tumor by comparatively high precision and offers a subsequent view to the radiologist.

23 citations

Proceedings ArticleDOI
01 Dec 2010
TL;DR: A reversible blind watermarking scheme to watermark a medical image with the patient's photograph in an invisible manner such that it is available to the respective physician only on extraction with a key is presented.
Abstract: The evolution of telemedicine has lead to the explosive increase in the exchange of medical images among remotely placed medical practitioners. As the medical images serve as valuable evidences for insurance claims and legal trials as well, it is required to ensure that these images are unaltered and they should be identified to be of that of an individual unambiguously. A recent research has shown that appending the patient's photograph with the medical image improves telediagnosis rather than just sending the medical image to a remote radiologist. To the contrary, major health policies have stipulated stringent rules regarding the disclosure of identity of the patients. This paper presents a reversible blind watermarking scheme to watermark a medical image with the patient's photograph in an invisible manner such that it is available to the respective physician only on extraction with a key.

22 citations

Journal ArticleDOI
TL;DR: A method namely co-occurrence matrix based algorithm for finding the number of clusters and silhouette index values that are used for cluster validation and the classification accuracy is evaluated by the parameters sensitivity, specificity, and accuracy.
Abstract: This paper proposes a hybrid technique for color image segmentation. First an input image is converted to the image of CIE L*a*b* color space. The color features "a" and "b" of CIE L*a*b* are then fed into fuzzy C-means (FCM) clustering which is an unsupervised method. The labels obtained from the clustering method FCM are used as a target of the supervised feed forward neural network. The network is trained by the Levenberg-Marquardt back-propagation algorithm, and evaluates its performance using mean square error and regression analysis. The main issues of clustering methods are determining the number of clusters and cluster validity measures. This paper presents a method namely co-occurrence matrix based algorithm for finding the number of clusters and silhouette index values that are used for cluster validation. The proposed method is tested on various color images obtained from the Berkeley database. The segmentation results from the proposed method are validated and the classification accuracy is evaluated by the parameters sensitivity, specificity, and accuracy.

22 citations


Cited by
More filters
Proceedings ArticleDOI
18 Mar 2016
TL;DR: This paper gives the survey about different classification techniques used for predicting the risk level of each person based on age, gender, Blood pressure, cholesterol, pulse rate using Naïve Bayes, KNN, Decision Tree Algorithm, Neural Network.
Abstract: Nowadays, health disease are increasing day by day due to life style, hereditary. Especially, heart disease has become more common these days, i.e. life of people is at risk. Each individual has different values for Blood pressure, cholesterol and pulse rate. But according to medically proven results the normal values of Blood pressure is 120/90, cholesterol is and pulse rate is 72. This paper gives the survey about different classification techniques used for predicting the risk level of each person based on age, gender, Blood pressure, cholesterol, pulse rate. The patient risk level is classified using datamining classification techniques such as Naive Bayes, KNN, Decision Tree Algorithm, Neural Network. etc., Accuracy of the risk level is high when using more number of attributes.

167 citations

Proceedings ArticleDOI
23 Jun 2017
TL;DR: This paper has analyzed prediction systems for Heart disease using more number of input attributes, which uses medical terms such as Gender, blood pressure, cholesterol like13 attributes to predict the likelihood of patient getting a Heart disease.
Abstract: The healthcare industry collects large amounts of Healthcare data, but unfortunately not all the data are mined which is required for discovering hidden patterns and effective decision making. We propose efficient genetic algorithm with the back propagation technique approach for heart disease prediction. This paper has analyzed prediction systems for Heart disease using more number of input attributes. The System uses medical terms such as Gender, blood pressure, cholesterol like13 attributes to predict the likelihood of patient getting a Heart disease.

108 citations

Journal ArticleDOI
01 Oct 2014-Micron
TL;DR: In this review, this work has categorized, evaluated, and discussed recently developed methods for leukocyte identification, and found their constraints.

97 citations

Journal ArticleDOI
TL;DR: An Internet of Things-based medical device for collecting patients’ heart details before and after heart disease is introduced and the HOBDBNN method and IoT-based analysis recognize heart disease with 99.03% accuracy with minimum time complexity.

92 citations