Scholarlink

Brain tumours are a leading cause of death worldwide, with high mortality rates. Early detection and treatment are vital for better patient outcomes. CT scans play a crucial role in diagnosis. Currently, deep learning-based solutions have been implemented to diagnose this disease early to prepare suitable treatment plans. Optimizers play a crucial role in deep neural networks, directly influencing the model’s accuracy. Deep learning, though parametric, aims to minimize assumptions. Gradient descent, a key optimization technique, is widely employed in neural networks and machine learning. However, traditional SGD and SGD with momentum faced challenges, prompting the development of adaptive learning optimizers like RMSprop, Adagrad, and Adam. This paper presents a comparative analysis of Adam, Nadam, AdamW, and Adamax optimizers using a Brain Computed Tomography (CT) dataset.

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

The Effect of Optimizers on the Classification Accuracy of Deep Neural Networks for Brain Tumor CT Scans

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Farrukh Hassan,
  • David Olayemi Alebiosu,
  • Yawar Abbas Bangash,
  • Samuel Mofoluwa Ajibade

摘要

Brain tumours are a leading cause of death worldwide, with high mortality rates. Early detection and treatment are vital for better patient outcomes. CT scans play a crucial role in diagnosis. Currently, deep learning-based solutions have been implemented to diagnose this disease early to prepare suitable treatment plans. Optimizers play a crucial role in deep neural networks, directly influencing the model’s accuracy. Deep learning, though parametric, aims to minimize assumptions. Gradient descent, a key optimization technique, is widely employed in neural networks and machine learning. However, traditional SGD and SGD with momentum faced challenges, prompting the development of adaptive learning optimizers like RMSprop, Adagrad, and Adam. This paper presents a comparative analysis of Adam, Nadam, AdamW, and Adamax optimizers using a Brain Computed Tomography (CT) dataset.