Title: Analysis of Medical Images using Machine Learning Techniques


Volume 9 Issue 1 Year 2019

Authors:

S.R. Dogiwal

dogiwal@gmail.com
Department of Computer Science & Engineering, Swami Keshvanand Institute of Technology, Management & Gramothan Jaipur-302017, (INDIA),

Mahesh Kumar Joshi

mjoshi293@gmail.com
Department of Computer Science & Engineering, Swami Keshvanand Institute of Technology, Management & Gramothan Jaipur-302017, (INDIA)

Pages: 11-15


Abstract:

In medical research and development, the computational domain provides support and calculative power to perform unique and innovative tasks. In computer field Image categorization is a huge and an important issue. In the world of medical science everyday doctors are encountered with various
types of image related problem. Pattern classification is a solution to categorize the image. In our human body brain play very important role, although due to the sudden growth of the anomalous tissues in the brain caused a very critical disease named as brain cancer. There are two different types of brain tumor such as benign and malignant. When cells and tissues are relatively grow slowly such type of cancer named as benign tumor which is termed as non cancerous tumor while in the malignant type of tumor, cells are growing very fast and causes serious harmful to the brain which causes death. There are so many techniques for extracting the medical images related to brain but MRI (Magnetic Resonance Imaging) is very powerful technique for extracting the digital images of internal slice of the brain. This information is very helpful for medical diagnosis and research purpose. To identification the tumor in human body firstly it goes through image acquisition process and then through the proper image segmentation technique finds out the tumor. The proposed work is motivated from the medical image analysis for medical dataset like brain tumor. In this research for image segmentation Otsu’s method is used and to select the appropriate features extraction principal component analysis is used. Gray level co-occurrence matrix is used to calculate some valuable texture parameters and finally using machine learning techniques kernel functions are used to calculate the highest accuracy with the help of support vector machine.

Keywords:
MRI, PCA, SVM, IDM, RMS