Scholarlink

This study focuses on classifying surgical forceps using deep learning to improve inventory management in medical settings. It targets specific types—such as Allis (curved and straight), Babcock, ring, field, Kelly, and Mixter forceps—using a custom photographic dataset. ResNet101 proved to be a feasible and accurate model for this task, showing strong potential for real-world application. This approach helps reduce equipment mismanagement and enhances operational efficiency. Future work will explore advanced architectures like Encoder-Decoder models, BLIP, YOLO, LSTM, and GRU to further improve classification accuracy and support deployment on devices with limited computational resources.

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

Convolutional Neural Networks for Accurate Medical Clamp Identification in Hospital Settings

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Yerlin Larissa Barahona Garcia,
  • Ariel Isaac Posada Barrera

摘要

This study focuses on classifying surgical forceps using deep learning to improve inventory management in medical settings. It targets specific types—such as Allis (curved and straight), Babcock, ring, field, Kelly, and Mixter forceps—using a custom photographic dataset. ResNet101 proved to be a feasible and accurate model for this task, showing strong potential for real-world application. This approach helps reduce equipment mismanagement and enhances operational efficiency. Future work will explore advanced architectures like Encoder-Decoder models, BLIP, YOLO, LSTM, and GRU to further improve classification accuracy and support deployment on devices with limited computational resources.