Investigation of the image processing algorithms for medical applications

• Main Author: Nurul Hidayah Abdul Manan
• Other Authors: Mohammed Yakoob Siyal
• Format: Final Year Project
• Language: English
• Published: 2017
• Subjects: DRNTU::Engineering::Electrical and electronic engineering
• Online Access: http://hdl.handle.net/10356/71434
• Institution: Nanyang Technological University

Biomedical image processing technique has been widely used in facilitating visualization analysis for medical images, for instance tumour. There are various types of tumours and different types of treatment required. This final year report covers the methods in order to enhance image by monitoring; using various methods to prevent misdiagnosis and the focus will be on brain tumour detection. Investigation and implementation of image processing algorithms is done for the detection of range and shape of tumour in brain MR images. Generally, Magnetic Resonance Imaging (MRI) scans or Computed Tomography (CT) scans are used to view the brain anatomy. However, the detailed pictures from MRI are more preferred modality to diagnose a tumour and it is commonly used as it does not contain any radiation as it uses magnetic field and radio waves [6][8]. Neurologists usually help to determine the type of MRI as there are different modalities of MRI available. For example, functional MRI (fMRI) shows information of location of area of the brain which are responsible for muscle movement and speech. The patient will be doing tasks which can cause changes in the brain and will be shown in the fMRI image. This test helps the surgeon to plan surgery and to ensure that while removing the tumour, there wouldn’t be any damage caused in the functional parts of the brain. [6] In addition, image segmentation was done to measure and visualize the brain’s anatomical structure. Different segmentation techniques are used (i.e. K-means clustering, Fuzzy-C means clustering and Hierarchical clustering) to evaluate the performance.