Automatic Detection of Tumor and Bleed in Magnetic Resonance Brain Images

Automatic Detection of Tumor and Bleed in Magnetic Resonance Brain Images

Jayanthi V. E., Jagannath Mohan, Adalarasu K.
ISBN13: 9781522551522|ISBN10: 1522551522|EISBN13: 9781522551539
DOI: 10.4018/978-1-5225-5152-2.ch015
Cite Chapter Cite Chapter

MLA

E., Jayanthi V., et al. "Automatic Detection of Tumor and Bleed in Magnetic Resonance Brain Images." Handbook of Research on Information Security in Biomedical Signal Processing, edited by Chittaranjan Pradhan, et al., IGI Global, 2018, pp. 291-303. https://doi.org/10.4018/978-1-5225-5152-2.ch015

APA

E., J. V., Mohan, J., & K., A. (2018). Automatic Detection of Tumor and Bleed in Magnetic Resonance Brain Images. In C. Pradhan, H. Das, B. Naik, & N. Dey (Eds.), Handbook of Research on Information Security in Biomedical Signal Processing (pp. 291-303). IGI Global. https://doi.org/10.4018/978-1-5225-5152-2.ch015

Chicago

E., Jayanthi V., Jagannath Mohan, and Adalarasu K. "Automatic Detection of Tumor and Bleed in Magnetic Resonance Brain Images." In Handbook of Research on Information Security in Biomedical Signal Processing, edited by Chittaranjan Pradhan, et al., 291-303. Hershey, PA: IGI Global, 2018. https://doi.org/10.4018/978-1-5225-5152-2.ch015

Export Reference

Mendeley
Favorite

Abstract

Brain tumor and intracerebral hemorrhage are major causes for death among the people. Brain tumor is the growth of abnormal cells multiplied in an uncontrolled manner in brain. Magnetic resonance imaging (MRI) technique plays a major role for analysis, diagnosis, and treatment planning of abnormalities in the brain. Bleed is detected manually by radiologists, but it is laborious, time-consuming, and error prone. The automatic detection method was performed to detect the tumor as well as bleed in brain under a single system. The proposed method includes image acquisition, pre-processing, patch extraction, feature extraction, convolutional neural network (CNN) classification, and fuzzy inference system (FIS) to detect the abnormality with reduced classification loss percentage. This chapter is compared with the existing system of tumor detection using convolution neural network based on certain features such as skewness, kurtosis, homogeneity, smoothness, and correlation.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.