Scholarlink

<p>Feed-forward neural networks with highly pre-processed inputs are proposed as an approach to automatically identify the presence of a wide variety of radioisotopes in complex samples. The process links characterised peaks from gamma spectroscopy to radionuclide emission lines. By inputting relational parameters between spectral peaks and decay data, the neural network selects between two possible solutions with 88% accuracy. The developed neural network demonstrates radionuclide insensitivity and can correctly identify radioisotopes that were not present in its training dataset with comparable accuracy.</p>

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

Developing Feed-forward neural networks to perform identification on a wide variety of radionuclides in gamma spectra

  • 期刊论文

  • 发表日期：

    2025年11月20日

查看全文
  • Jay Wroe-Brown,
  • Caroline Shenton-Taylor

摘要

Feed-forward neural networks with highly pre-processed inputs are proposed as an approach to automatically identify the presence of a wide variety of radioisotopes in complex samples. The process links characterised peaks from gamma spectroscopy to radionuclide emission lines. By inputting relational parameters between spectral peaks and decay data, the neural network selects between two possible solutions with 88% accuracy. The developed neural network demonstrates radionuclide insensitivity and can correctly identify radioisotopes that were not present in its training dataset with comparable accuracy.