Scholarlink

This study suggests overcoming the limitation by using a deep-learning model designed for detecting subtle cancerous areas, where the input data used are ultrasound images, which are pre-processed, augmented, and split into different sets for training and testing purposes. A classifier layer has also been included to improve the model's functionality. The findings demonstrate that the CNN architectures for deep learning, ResNet50 and VGG16, have achieved accuracy rates of 94% and 87%, respectively. Breast cancer diagnosis currently relies on traditional biopsy methods with the potential to miss up to 10–30 percent of early-stage cancer cases due to the difficulty of detecting small cancerous regions.

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

Breast Cancer Detection Using Deep Learning CNN Architectures

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Chelluboina Rajesh,
  • Anagha Harshini,
  • Garikipati Jyotsna

摘要

This study suggests overcoming the limitation by using a deep-learning model designed for detecting subtle cancerous areas, where the input data used are ultrasound images, which are pre-processed, augmented, and split into different sets for training and testing purposes. A classifier layer has also been included to improve the model's functionality. The findings demonstrate that the CNN architectures for deep learning, ResNet50 and VGG16, have achieved accuracy rates of 94% and 87%, respectively. Breast cancer diagnosis currently relies on traditional biopsy methods with the potential to miss up to 10–30 percent of early-stage cancer cases due to the difficulty of detecting small cancerous regions.