26 June 2017 Measuring the volume of brain tumour and determining its location in T2-weighted MRI images using hidden Markov random field: expectation maximization algorithm
Author Affiliations +
Proceedings Volume 10335, Digital Optical Technologies 2017; 103351V (2017) https://doi.org/10.1117/12.2270132
Event: SPIE Digital Optical Technologies, 2017, Munich, Germany
Abstract
A brain tumour is an abnormal growth of tissue in the brain. Most tumour volume measurement processes are carried out manually by the radiographer and radiologist without relying on any auto program. This manual method is a timeconsuming task and may give inaccurate results. Treatment, diagnosis, signs and symptoms of the brain tumours mainly depend on the tumour volume and its location. In this paper, an approach is proposed to improve volume measurement of brain tumors as well as using a new method to determine the brain tumour location. The current study presents a hybrid method that includes two methods. One method is hidden Markov random field - expectation maximization (HMRFEM), which employs a positive initial classification of the image. The other method employs the threshold, which enables the final segmentation. In this method, the tumour volume is calculated using voxel dimension measurements. The brain tumour location was determined accurately in T2- weighted MRI image using a new algorithm. According to the results, this process was proven to be more useful compared to the manual method. Thus, it provides the possibility of calculating the volume and determining location of a brain tumour.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Mohd. Zubir Mat Jafri, Hayder Saad Abdulbaqi, Kussay N. Mutter, Iskandar Shahrim Mustapha, Ahmad Fairuz Omar, "Measuring the volume of brain tumour and determining its location in T2-weighted MRI images using hidden Markov random field: expectation maximization algorithm", Proc. SPIE 10335, Digital Optical Technologies 2017, 103351V (26 June 2017); doi: 10.1117/12.2270132; https://doi.org/10.1117/12.2270132
PROCEEDINGS
5 PAGES


SHARE
Back to Top