Automated Classification and Annotation of Computed Tomography Brain Images

Citation

Tong, Hau Lee (2015) Automated Classification and Annotation of Computed Tomography Brain Images. PhD thesis, Multimedia University.

Full text not available from this repository.

Abstract

Brain hemorrhage detection is clinically crucial for the patients having head trauma and neurological disturbances. Early finding and accurate diagnosis of the brain abnormalities is one of the key contributions for the execution of the successful therapy and proper treatment. Multi-slice Computed Tomograph (CT) scans are widely employed in today’s examination of head traumas due to its effectiveness to disclose some abnormalities such as brain hemorrhages and so on. However, radiologists have to manually analyse the CT slices for the presence of brain hemorrhages. Due to the large volume of CT scan examinations, it is important to develop a computerised system that can assist the radiologists to automatically detect the presence of the brain abnormalities as well as automatically retrieve the images. This thesis presents an automated annotation and classification of the CT brain images. The main objective is to propose a new methodology to annotate and classify the different types of brain hemorrhages which are intra-axial, subdural and extradural hemorrhages. Besides, this thesis also aims to evaluate and investigate the effectiveness and suitability of different segmentation and classification techniques as well as introduce the new features for the classification.

Item Type: Thesis (PhD)
Additional Information: Call No.: R857.O6 T66 2015
Uncontrolled Keywords: Imaging systems in medicine
Subjects: R Medicine > R Medicine (General) > R856-857 Biomedical engineering. Electronics. Instrumentation
Divisions: Faculty of Computing and Informatics (FCI)
Depositing User: Ms Nurul Iqtiani Ahmad
Date Deposited: 12 Jul 2017 03:46
Last Modified: 12 Jul 2017 03:46
URII: http://shdl.mmu.edu.my/id/eprint/6854

Downloads

Downloads per month over past year

View ItemEdit (login required)