Brain segmentation of T1-weighted MRI brain images / Zalikha Farhanah Zainal Abidin
The Magnetic Resonance Imaging (MRI) scan is one of the technologies that offer prediction on abnormality inside the human skull and considered as the best way to detect tumour. Segmentation for the brain tumour is important since it will separate every tissue and any other elements that have been c...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://ir.uitm.edu.my/id/eprint/14604/ http://ir.uitm.edu.my/id/eprint/14604/1/TD_ZALIKHA%20FARHANAH%20ZAINAL%20ABIDIN%20CS%2015_5.pdf |
| Summary: | The Magnetic Resonance Imaging (MRI) scan is one of the technologies that offer prediction on abnormality inside the human skull and considered as the best way to detect tumour. Segmentation for the brain tumour is important since it will separate every tissue and any other elements that have been captured by the MRI scan. The brain tumour can be detected manually by the human experts. However, the expert might have
false judgement in detecting the tumour since brain tumour detection is a tedious and difficult task to do. Moreover, there is certain point where the MRI scan gives a vague image, then the use of gadolinium, a contrast agent are needed. The use of the contrast agent will give harmful side effect to the patients. Therefore, the solutions of brain tumour detection without the use of contrast agent are needed. It is important to detect the existence of tumour at early stage since it can lead to death.n However manual recognition is time consuming. The tumour needs its treatment as soon as possible. The brain tumour detection using MRI plays a major role in brain tumour detection. This project is implemented skull stripping process that will helps future work of brain tumour detection. This project will used double thresholding method and morphology method to segment the brain images and using 30 images of T1-weighted MRI brain
images as the tested data. This project accuracy results is considered acceptable with 92.438%. This project will contribute to brain tumour detection area where the proposed skull stripping methods can be used as a preliminary study in other brain tumour
detection application. |
|---|