分割
卷积神经网络
人工智能
计算机科学
模式识别(心理学)
图像分割
人工神经网络
数据集
深度学习
脑瘤
病理
医学
作者
Senthilkumar Rajendran,Suresh Kumar Rajagopal,Tamilvizhi Thanarajan,K. Shankar,Sachin Kumar,Najah Alsubaie,Mohamad Khairi Ishak,Samih M. Mostafa
出处
期刊:IEEE Access
[Institute of Electrical and Electronics Engineers]
日期:2023-01-01
卷期号:11: 64758-64768
被引量:3
标识
DOI:10.1109/access.2023.3288017
摘要
Segmenting brain tumors automatically using MR data is crucial for disease investigation and monitoring.Due to the aggressive nature and diversity of gliomas, well-organized and exact segmentation methods are used to classify tumors intra-tumorally.The proposed technique uses a Gray Level Cooccurrence matrix extraction of features approach to strip out unwanted details from the images.In comparison with the current state of the art, the accuracy of brain tumor segmentation was significantly improved using Convolutional Neural Networks, which are frequently used in the field of biomedical image segmentation.By merging the results of two separate segmentation networks, the proposed method demonstrates a major but simple combinatorial strategy that, as a direct consequence, yields much more precise and complete estimates.A U-Net and a Three-Dimensional Convolutional Neural Network.These networks are used to break up images into their component parts.Following that, the prediction was constructed using two distinct models that were combined in a number of ways.In comparison to existing state-of-the-art designs, the proposed method achieves the mean accuracy (%) of 99.40, 98.46, 98.29, precision (%) of 99.41, 98.51, 98.35, 98.29, 98.46 and sensitivity (%) of 99.39, 98.41, 98.25 for the whole tumor, enhanced tumor, tumor core on the validation set, respectively.
科研通智能强力驱动
Strongly Powered by AbleSci AI