Edge Methods and Ensemble Classifier for Leukaemia Diagnosis
In information technology, health informatics have been qualified as the prominent province. This skilled person is trackable to such a sophisticated creation of medical care informatics in a short time span, it is possible to diagnose multiple disorders nowadays. There is one disease dub leukaemia in relation to complaints which helps to recognize by manipulating the various types. In recent researches, it is recognized through classification via the usefulness of incorporating Fuzzy Neural Network (FNN) as well as Enhanced Support Vector Machine (ESVM) classifiers named as FNESVM. But here it eliminates the edges of the nuclei for whole images of the nuclei imperfectly and the single classifiers helps in performing the classification task, which isn’t enhances the accuracy of the classifier. In this work a new method was introduced to eliminate the problems through enforcing the edge detection and edge enhancement methods. In proposed work, the algorithm selection experiment assumes three classification algorithms specifically Adaptive Neuro Fuzzy Inference System (ANFIS), RBF network (RBFNETS), Online Learning with Support Vector Machine (OLSVM). An average rule helps in integrating the ensemble of classifier. The proposed work experimental output is distinguished with the leukaemia detection techniques. The robustness of the proposed techniques were explained by proceeding the qualitative and quantitative analysis on leukaemia image collections.
Keywords: Health informatics, Classification, Leukaemia diagnose, bone-marrow, Edge detection, visibility and Ensemble of classifier.