Scholarlink

The brain tumor is a set of cells with uncontrolled development of brain cells. Early identification of tumors is essential for diagnosis, treatment planning, and follow-up after cancer treatments. There are several types, forms, stages, and properties of brain tumors. Therefore, manually detecting these tumors is challenging, time-consuming, and prone to mistakes. Consequently, there is currently a need for automated, highly precise computer-assisted diagnosis. We propose in this paper four deep learning CNN models that have a high accuracy rate for classification, an MR image dataset was used to make predictions about the presence of brain tumors. Any errors in spelling, grammar, or punctuation have been corrected. To achieve this, four models were used, the models We achieved high accuracy with the following results: InceptionV3 (96.02%), ResNet50 (97.25%), Xception (86%), and DenseNet121 (54%).

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

Predicting Brain Tumors Using Ensembles of Deep Learning Models

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Khadija Elaissaoui,
  • Jalal Rabbah,
  • Mohammed Ridouani

摘要

The brain tumor is a set of cells with uncontrolled development of brain cells. Early identification of tumors is essential for diagnosis, treatment planning, and follow-up after cancer treatments. There are several types, forms, stages, and properties of brain tumors. Therefore, manually detecting these tumors is challenging, time-consuming, and prone to mistakes. Consequently, there is currently a need for automated, highly precise computer-assisted diagnosis. We propose in this paper four deep learning CNN models that have a high accuracy rate for classification, an MR image dataset was used to make predictions about the presence of brain tumors. Any errors in spelling, grammar, or punctuation have been corrected. To achieve this, four models were used, the models We achieved high accuracy with the following results: InceptionV3 (96.02%), ResNet50 (97.25%), Xception (86%), and DenseNet121 (54%).