Implementation of Brain Tumor Segmentation Using CNN Deep Learning Algorithm

Barsiya, Tarang Kumar; Bhargava, Lakshita; Agrawal, Suchitra; Tiwari, Aruna; Saxena, Amit

doi:10.1007/978-981-16-2712-5_59

Cited by 1 publication

(1 citation statement)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A subset of machine learning (ML) called deep learning (DL) is focused on hierarchical feature learning and learning data representation [4]. A system of many layers of nonlinear processing identities is used by CNN to extract features [1]. As we go deeper into the network, data abstraction is aided by the fact that the input of one sequential layer becomes the output of the preceding one.…”

Section: Introductionmentioning

confidence: 99%

Multi-class Brain Tumor Classification and Segmentation using Hybrid Deep Learning Network Model

Kumar

Bonthu

Meghana

et al. 2023

SCPE

View full text Add to dashboard Cite

Brain tumor classification is a significant task for evaluating tumors and selecting the type of treatment as per their classes. Brain tumors are diagnosed using multiple imaging techniques. However, MRI is frequently utilized since it provides greater image quality and uses non-ionizing radiation. Deep learning (DL) is a subfield of machine learning and recently displayed impressive performance, particularly in segmentation and classifying problems. Based on convolutional neural network (CNN), a Hybrid Deep Learning Network (HDLN) model is proposed in this research for classifying multiple types of brain tumors including glioma, meningioma, and pituitary tumors. The Mask RCNN is used for brain tumor classification. We used a squeeze-and-excitation residual network (SE-ResNet) for brain tumor segmentation, which is a residual network (ResNet) with a squeeze-and-excitation block. A publicly available research dataset is used for testing the proposed model for experiment analysis and it obtained an overall accuracy of 98.53%, 98.64% sensitivity and 98.91% specificity. In comparison to the most advanced classification models, the proposed model obtained the best accuracy. For multi-class brain tumor diseases, the proposed HDLN model demonstrated its superiority to the existing approaches.

show abstract

Section: Introductionmentioning

confidence: 99%