Scholarlink

<p>Brain tumors are a major cause of neurological morbidity and mortality, emphasizing the critical need for early and precise diagnosis to facilitate effective treatment planning. Conventional diagnostic methods can be invasive or time-consuming. Therefore, this study explores a robust hybrid system for classifying brain tumors in magnetic resonance imaging (MRI) images by integrating machine learning and deep learning techniques. This work utilizes brightness-preserved bi-histogram equalization (BBHE) to enhance contrast of MRI images. Then, a lightweight convolutional neural network (CNN) architecture was used to extract both local and global features from the enhanced images. Further, ReliefF algorithm is then applied to select the most relevant features for reducing redundancy and improving classification accuracy. Finally, classification is performed using support vector machines, K-nearest neighbors (KNN), artificial neural networks (ANN), and an ensemble model. Evaluation is conducted on a publicly available Kaggle dataset comprising four classes: glioma, meningioma, pituitary, and normal. Experimental results demonstrate that our model achieves an accuracy of 98.79%, outperforming existing approaches. This study proposes a novel method integrating BBHE pre-processing, CNN feature extraction, ReliefF feature selection, and an ensemble classifier for accurate brain tumor classification. The approach shows promise for enhancing diagnostic efficiency and improving patient outcomes.</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

A unified hybrid framework for accurate and early brain tumor diagnosis through MRI analysis

  • 期刊论文

  • 发表日期：

    2025年6月16日

查看全文
  • Srinivas Babu Gottipati,
  • Gowri Thumbur

摘要

Brain tumors are a major cause of neurological morbidity and mortality, emphasizing the critical need for early and precise diagnosis to facilitate effective treatment planning. Conventional diagnostic methods can be invasive or time-consuming. Therefore, this study explores a robust hybrid system for classifying brain tumors in magnetic resonance imaging (MRI) images by integrating machine learning and deep learning techniques. This work utilizes brightness-preserved bi-histogram equalization (BBHE) to enhance contrast of MRI images. Then, a lightweight convolutional neural network (CNN) architecture was used to extract both local and global features from the enhanced images. Further, ReliefF algorithm is then applied to select the most relevant features for reducing redundancy and improving classification accuracy. Finally, classification is performed using support vector machines, K-nearest neighbors (KNN), artificial neural networks (ANN), and an ensemble model. Evaluation is conducted on a publicly available Kaggle dataset comprising four classes: glioma, meningioma, pituitary, and normal. Experimental results demonstrate that our model achieves an accuracy of 98.79%, outperforming existing approaches. This study proposes a novel method integrating BBHE pre-processing, CNN feature extraction, ReliefF feature selection, and an ensemble classifier for accurate brain tumor classification. The approach shows promise for enhancing diagnostic efficiency and improving patient outcomes.