Brain Tumor Segmentation Using Fully Convolutional Networks from Magnetic Resonance Imaging

Zhang, Changjiang; Fang, Mingchao; Nie, Huanhuan

doi:10.1166/jmihi.2018.2502

Cited by 4 publications

(2 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With the advancement of deep learning, medical image segmentation has made significant progress. For instance, Zhang et al [4] applied a fully convolutional network(FCN) model to segment brain MRI images [2] . They initially segmented the tumor regions in the images and further trained the network to achieve more accurate segmentation results.…”

Section: Introductionmentioning

confidence: 99%

Attentional mechanisms for multiscale aggregate U-Net brain tumor image segmentation

Yuan

2024

Fourth International Conference on Sensors and Information Technology (ICSI 2024)

View full text Add to dashboard Cite

In response to the need for brain tumor image segmentation, which involves segmenting it into three different regions based on the degree of tumor lesions, this paper addresses several issues with the Unet model, such as limited feature extraction accuracy and unclear region segmentation. To tackle these challenges, we propose a multi-scale aggregation Unet segmentation algorithm that incorporates an attention mechanism. The proposed algorithm follows a two-step process: first, it conducts feature extraction and up-sampling after each layer's down-sampling in the Unet architecture, and then iteratively aggregates the feature map with the up-sampled map. Second, we introduce residual blocks in the encoding stage to address the issue of gradient vanishing during downsampling. Finally, we incorporate channel attention mechanisms across layers in the decoding stage to direct the network's focus on the relevant regions, thus improving the segmentation accuracy of lesion regions. We conducted simulation experiments using the BraTs2021 training set, achieving Dice coefficients of 0.927, 0.861, and 0.867 for the segmentation of the whole tumor, core tumor, and enhanced tumor images, respectively. These results surpass those of most existing brain tumor segmentation models.

show abstract

Section: Introductionmentioning

confidence: 99%

Attentional mechanisms for multiscale aggregate U-Net brain tumor image segmentation

Yuan

2024

Fourth International Conference on Sensors and Information Technology (ICSI 2024)

View full text Add to dashboard Cite

show abstract

“…The conditional random fields (CRF) technique can also be used to detect BT with minimal computation time [6]. Followed by, convolutional neural network (CNN) models find useful as a supervised classification model in which distinct data undergo segmentation without requiring distributed parametric hypothesis [7].…”

Section: Introductionmentioning

confidence: 99%

An Optimal Tsallis Entropy Based Segmentation With Deep Extreme Learning Machine for Medical Image Classification Using Mri Images

Deeparani¹,

Sudhakar²

2021

INDJCSE

View full text Add to dashboard Cite

Medical image classification becomes an essential process in the healthcare sector for decision making using medicinal images. At the same time, brain tumor (BT) is considered the deadliest disease over the globe and it affects the brain tissues. So, it is needed to diagnose BT at the preliminary stages to enhance the survival rate of the patients. The goal of this paper is to derive an effective segmentation and classification technique for BT diagnosis. This paper designs an optimal Tsallis entropy based segmentation with deep extreme learning machine (OTSE-DELM) model for medical image classification using MRI images. The presented model involves multi-level thresholding based image segmentation using Tsallis entropy and the optimal threshold values are chosen by black widow optimization (BWO) algorithm. Besides, the presented model encompasses directional local extrema patterns (DLEP) based feature extractor. In addition, the OTSE-DELM model involves DELM based classification model to allocate the proper class labels of the test MRI images. A wide range of simulations was performed to showcase the betterment of the OTSE-DELM model and the outcomes are investigated under distinct dimensions. The experimental values make sure that the OTSE-DELM model is found to be superior to other methods by providing an increased sensitivity of 98.34%, specificity of 98.98%, and accuracy of 98.76%.

show abstract

A New Approach for Brain Tumor Segmentation and Classification Based on Score Level Fusion Using Transfer Learning

et al. 2019

View full text Add to dashboard Cite

Brain Tumor Segmentation Using Fully Convolutional Networks from Magnetic Resonance Imaging

Cited by 4 publications

References 0 publications

Attentional mechanisms for multiscale aggregate U-Net brain tumor image segmentation

Attentional mechanisms for multiscale aggregate U-Net brain tumor image segmentation

An Optimal Tsallis Entropy Based Segmentation With Deep Extreme Learning Machine for Medical Image Classification Using Mri Images

A New Approach for Brain Tumor Segmentation and Classification Based on Score Level Fusion Using Transfer Learning

Contact Info

Product

Resources

About