Scholarlink

Twenty years ago, the use of deep learning techniques for 3D segmentation of medical images has become increasingly popular due to their effectiveness in recognizing complex patterns and anatomical structures from training data, such as segmentation of brain tumors in 3D Magnetic Resonance Images (MRIs). This segmentation is an essential step in the process of early diagnosis and treatment evaluation. When performed manually, it often takes a lot of time, and the segmentation performance varies depending on the experience of the doctors. This leads to the need for a fully automatic method for 3D segmentation of brain tumors. To achieve this goal, we use artificial intelligence techniques to analyze three-dimensional medical images (IRM). These techniques significantly improve the detection and segmentation of brain tumors by reducing segmentation errors and providing more accurate information about the size, shape, and location of tumors. This will have a major impact in clinical domains, helping doctors make more informed decisions about treatments and improving outcomes for patients with brain tumors. In this paper, we put out a plan using a group of three learning models (3D CNN, 3D U-Net, and 3D V-net). We trained and evaluated on the BraTS dataset, which has 1266 brain tumor cases.

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

Ensemble Learning for 3D Brain Tumor Segmentation Using Deep Learning Techniques on the BraTS Dataset

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Maryem Belkaid,
  • Abir Maali,
  • El Arbi Abdellaoui Alaoui,
  • Achraf Berrajaa,
  • Mostafa Merras

摘要

Twenty years ago, the use of deep learning techniques for 3D segmentation of medical images has become increasingly popular due to their effectiveness in recognizing complex patterns and anatomical structures from training data, such as segmentation of brain tumors in 3D Magnetic Resonance Images (MRIs). This segmentation is an essential step in the process of early diagnosis and treatment evaluation. When performed manually, it often takes a lot of time, and the segmentation performance varies depending on the experience of the doctors. This leads to the need for a fully automatic method for 3D segmentation of brain tumors. To achieve this goal, we use artificial intelligence techniques to analyze three-dimensional medical images (IRM). These techniques significantly improve the detection and segmentation of brain tumors by reducing segmentation errors and providing more accurate information about the size, shape, and location of tumors. This will have a major impact in clinical domains, helping doctors make more informed decisions about treatments and improving outcomes for patients with brain tumors. In this paper, we put out a plan using a group of three learning models (3D CNN, 3D U-Net, and 3D V-net). We trained and evaluated on the BraTS dataset, which has 1266 brain tumor cases.