Scholarlink

<p>Early brain tumor detection can avert millions of fatalities and the improvement of survival of patients can be attained by the use of Magnetic Resonance Imaging (MRI), which aids in further treatment. Most of the existing techniques in brain tumor detection are less accurate, time-consuming, and less efficient. Hence, automatic classification is performed for the effective classification of tumors. In this research, the Egocentric Critter Fish Optimization-based Deep Convolutional Neural Network (ECFO-DCNN) is proposed for the effective classification of brain tumors. The effective classification is attained using the ECFO-based segmentation and Binary, Directional, Orientational, and Entropy (BDOE) features that boost the convergence and reduce the classifier’s required computational time. The effectiveness of the method is established through the comparison with existing approaches, where the ECFO-DCNN method achieved 90.43% accuracy, 80.65% sensitivity, and 95.32% specificity using the Figshare brain tumor dataset.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

ECFO-DCNN: Egocentric Critter Fish Optimization Enabled Deep Convolutional Neural Network for Brain Tumor Classification

  • 期刊论文

  • 发表日期：

    2025年4月1日

查看全文
  • Mahesh P. Potadar,
  • Raghunath S. Holambe,
  • Rajan H. Chile

摘要

Early brain tumor detection can avert millions of fatalities and the improvement of survival of patients can be attained by the use of Magnetic Resonance Imaging (MRI), which aids in further treatment. Most of the existing techniques in brain tumor detection are less accurate, time-consuming, and less efficient. Hence, automatic classification is performed for the effective classification of tumors. In this research, the Egocentric Critter Fish Optimization-based Deep Convolutional Neural Network (ECFO-DCNN) is proposed for the effective classification of brain tumors. The effective classification is attained using the ECFO-based segmentation and Binary, Directional, Orientational, and Entropy (BDOE) features that boost the convergence and reduce the classifier’s required computational time. The effectiveness of the method is established through the comparison with existing approaches, where the ECFO-DCNN method achieved 90.43% accuracy, 80.65% sensitivity, and 95.32% specificity using the Figshare brain tumor dataset.