A Deep Learning Architecture for Meningioma Brain Tumor Detection and Segmentation

• Published in: Journal of cancer prevention Vol. 27; no. 3; pp. 192 - 198
• Main Authors: Anita, John Nisha, Kumaran, Sujatha
• Format: Journal Article
• Language: English
• Published: Korean Society of Cancer Prevention 30.09.2022
• Subjects: Original
• ISSN: 2288-3649; 2288-3657
• DOI: 10.15430/JCP.2022.27.3.192

The meningioma brain tumor detection and segmentation method is a complex process due to its low intensity pixel profile. In this article, the meningioma brain tumor images were detected and tumor regions were segmented using a convolutional neural network (CNN) classification approach. The source brain MRI images were decomposed using the discrete wavelet transform and these decomposed sub bands were fused using an arithmetic fusion technique. The fused image was data augmented in order to increase the sample size. The data augmented images were classified into either healthy or malignant using a CNN classifier. Then, the tumor region in the classified meningioma brain image was segmented using an connection component analysis algorithm. The tumor region segmented meningioma brain image was compressed using a lossless compression technique. The proposed method stated in this article was experimentally tested with the sets of meningioma brain images from an open access dataset. The experimental results were compared with existing methods in terms of sensitivity, specificity and tumor segmentation accuracy.