相关实验视频

Updated: Sep 11, 2025

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

可解释的深度学习解锁了医疗放射性同位素生产的高准确性预测

¹University of Chinese Academy of Sciences, Beijing, China; European Organization for Nuclear Research (CERN), Geneva, Switzerland.

Applied radiation and isotopes : including data, instrumentation and methods for use in agriculture, industry and medicine

|August 18, 2025

概括

我们使用贝叶斯优化的深度神经网络开发了一个人工智能模型,以准确预测必要的医疗放射性同位素生产的核反应截面. 这种方法提高了准确性和可解释性,优化了用于核医学的放射性同位素生成.

科学领域:

核物理核物理核物理
医疗放射性同位素生产医疗放射性同位素生产
科学中的人工智能.

背景情况:

准确的核反应截面数据对于高效的医疗放射性同位素生产至关重要.
基于物理学的传统模型由于不确定性和稀少的实验数据而存在局限性.
优化诊断和治疗放射性同位素的生产策略是一个重大挑战.

研究的目的:

为 (p,2n) 反应截面开发一个高度准确的预测框架.
改进重要的医疗放射性同位素如Sc-47,In-111,I-124和Tm-165.5的生产策略.
为影响横截面预测的因素提供可解释的见解.

主要方法:

利用贝叶斯优化的深度神经网络进行横截面预测.
在国际原子能机构 (IAEA) 数据库的评估数据上进行训练和验证的模型.
为了模型的可解释性,使用了SHAP (夏普利添加式解释) 分析.

主要成果:

在预测 (p,2n) 反应截面方面取得了非常高的准确性,Pearson相关系数 (R) 为0.9997.
通过使用默认参数,超过了TALYS-2.0核反应代码 (R=0.9783).
确定了射弹能量和核结构特征作为关键预测因素.

结论:

数据驱动,可解释的AI模型可以克服核数据评估中的挑战.
开发的框架提供了一个强大的工具,用于优化循环基放射性同位素的生产.
这项工作通过改进放射性同位素生成,推动了核医学领域的发展.

关键词:

深度学习是一种深度学习.人工智能的解释性医疗放射性同位素的生产核反应的横截面.

更多相关视频

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

相关概念视频

Positron Emission Tomography

Positron Emission Tomography

Positron emission tomography (PET) is a medical imaging technique involving radiopharmaceuticals — substances that emit short-lived radiation. Although the first PET scanner was introduced in 1961, it took 15 more years before radiopharmaceuticals were combined with the technique and revolutionized its potential.
One of the main requirements of a PET scan is a positron-emitting radioisotope, which is produced in a cyclotron and then attached to a substance used by the part of the body...

Isotopes and Radioisotopes

Isotopes and Radioisotopes

In the early 1900s, English chemist Frederick Soddy realized that an element could have atoms with different masses that were chemically indistinguishable. These different types are called isotopes — atoms of the same element that differ in mass. Isotopes differ in mass because they have different numbers of neutrons but are chemically identical because they have the same number of protons. Soddy was awarded the Nobel Prize in Chemistry in 1921 for this discovery.
An isotope containing...