Scholarlink

The article reports on the use of deep neural networks (DNNs) in image analysis to assess the analgesic activity of drugs in laboratory mice. The authors developed a two-plate analgesiameter with a camera to record time spent by animals on plates set at different temperatures. The results were analyzed using classical algorithms and CNNs such as ResNet and GoogLeNet, as well as the authors’ own CNN. The author’s own CNN network achieved the highest accuracy (97.63%) and balanced TPR, TNR, PPV, and NPV. The method outperformed classical algorithms, minimizing errors and stress on the animals tested. Further analysis of the movement behavior of mice is planned in the future.

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

Deep Neural Networks as a Tool for Evaluating the Analgesic Activity of Drugs

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Robert Sałat,
  • Mirosław Oleksy,
  • Michał Awtoniuk,
  • Miłosz Worwa,
  • Jędrzej Trajer,
  • Kinga Sałat,
  • Linh Tran

摘要

The article reports on the use of deep neural networks (DNNs) in image analysis to assess the analgesic activity of drugs in laboratory mice. The authors developed a two-plate analgesiameter with a camera to record time spent by animals on plates set at different temperatures. The results were analyzed using classical algorithms and CNNs such as ResNet and GoogLeNet, as well as the authors’ own CNN. The author’s own CNN network achieved the highest accuracy (97.63%) and balanced TPR, TNR, PPV, and NPV. The method outperformed classical algorithms, minimizing errors and stress on the animals tested. Further analysis of the movement behavior of mice is planned in the future.