Scholarlink

Magnetic resonance imaging (MRI) is essential for the management and diagnosis of neurological disorders. This investigation introduces an integrated methodology that integrates the Extra Trees classifier with Discrete Wavelet Transform (DWT)-based denoising to improve the classification of brain MRI images into normal and aberrant categories. The research employs a dataset of 800 MRI images, of which 408 are classified as normal and 392 as aberrant. Our classification performance is significantly enhanced as a result of the significant improvements in our preprocessing techniques, which effectively resolve challenges related to noise and variability. The accuracy of the proposed method is 99% when denoising is implemented, as opposed to 97% when denoising is not implemented. The potential of the integrated approach as a dependable diagnostic instrument for detecting brain abnormalities in MRI images is demonstrated by these results, which underscore its effectiveness.

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

Optimizing Brain MRI Classification: A Novel Approach Integrating DWT-Based Denoising with Extra Trees Ensemble for Superior Diagnostic Accuracy

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Mohamed El Ghmary,
  • Ziani Said,
  • M. Oubakalla

摘要

Magnetic resonance imaging (MRI) is essential for the management and diagnosis of neurological disorders. This investigation introduces an integrated methodology that integrates the Extra Trees classifier with Discrete Wavelet Transform (DWT)-based denoising to improve the classification of brain MRI images into normal and aberrant categories. The research employs a dataset of 800 MRI images, of which 408 are classified as normal and 392 as aberrant. Our classification performance is significantly enhanced as a result of the significant improvements in our preprocessing techniques, which effectively resolve challenges related to noise and variability. The accuracy of the proposed method is 99% when denoising is implemented, as opposed to 97% when denoising is not implemented. The potential of the integrated approach as a dependable diagnostic instrument for detecting brain abnormalities in MRI images is demonstrated by these results, which underscore its effectiveness.