2019 · 16 citations · 14 references
MedicineComputer ScienceOncologyDiagnostic ImagingMachine LearningNeuro-oncologyBrain TumorsClassificationMagnetic Resonance ImagingSupport Vector MachinesTumor ClassificationTumor DetectionAutomatic SegmentationBenign Tumor ClassificationBrain Tumor DiagnosisGray-level Co-occurrence MatricesLesion ClassificationMedical Image ClassificationSupport Vector ClassificationSupport Vector Machine Classification
Now a day's brain tumor is the most critical disease in human being, it creates the major issues related to the human nervous system. The tumor is a random growth of cells in the brain. Its detection using segmentation is one of the most challenging processes in medical field. The manual process of image understanding consumes lot of time and it may be the error prone task. The Magnetic Resonance Imaging (MRI) technique in one of the important method to get vital information about the soft (brain part) and hard (skull) tissues; which helps in the diagnosis of brain tumor. To solve this problem, the automatic segmentation and classification technique is very helpful and consumes less time. In the research work classification of tumor in two types i.e. malignant and benign is proposed based on MRI images. Gray Level Co-occurrence Matrix (GLCM) feature extraction method is used and feature dataset is prepared. Support vector machine (SVM) and KNN classifiers are used to classify tumors in malignant and benign type. In this paper hybrid approach of classifier is proposed, it is the combination of SVM and KNN classifiers to improve efficiency.
14
MRI brain cancer classification using hybrid classifier (SVM-KNN)
Ketan Machhale, Hari Babu Nandpuru, Vivek Kapur et al. · 2015 · 148 citations
Medicine, Computer Science, Oncology +17