Accurate and Efficient Diagnosis of Brain Tumour Disease using Kernel Support Vector Machine

Brain tumor often recognized to be a generation of abnormal cells within brain. Brain tumor needs to be diagnosed accurately for the proper decision making about the disease occurrence, so that appropriate treatment on time can be provided to the patients. However analyzing and diagnosing the brain tumor from the imbalanced data set would be more difficult process which needs to handled very carefully for the accurate brain tumor diagnosis. Physiological data of patients are the primary and vital entities in healthcare big data analytic. Hence, valid raw data must be collected with an efficient manner in a medical environment. In the existing research work, Integrated feature Selection and Ensemble Classification (IFSEC) is introduced which attempts to accurately predict the brain tumor disease. However, handling large volume of data in the single machine would lead to inaccurate and wrong diagnosis of brain tumor disease which needs to be handled with more concern for the accurate output. This is resolved in the proposed research method by introducing novel method namely Accurate Prediction of Brain Tumor Disease from Big Data Framework (APBTD-BDF). To make ease of large volume of data handling process, in the proposed research work, content aware partitioning is introduced which attempts to divide the health care big data into multiple partitions which can be scheduled on different machines for the classification purpose. To improve the classification accuracy in this research low variance filtering approach is introduced for the selection of the more important features. Classification is done by using Kernel Support Vector Machine (KSVM) approach whose performance is tested based 10 fold cross validation technique. The overall research method is implemented in the MATLAB simulation environment from which it can be proved that the proposed research method can effectively handle the large volume of data and it can predict the brain tumor disease more accurately.