Genetic algorithm search for the worst‐case MRI RF exposure for a multiconfiguration implantable fixation system modeled using artificial neural networks

Genetic algorithm search for the worst‐case MRI RF exposure for a multiconfiguration implantable fixation system modeled using artificial neural networks
复制标题

遗传算法搜索使用人工神经网络建模的多配置植入固定系统的最坏情况 MRI 射频暴露

DOI:
10.1002/mrm.28319
复制
发表时间:
2020
影响因子:
3.3
通讯作者:
Chen, Ji
Chen, Ji
中科院分区:
医学3区
文献类型:
--
作者:
Zheng, Jianfeng;Lan, Qianlong;Kainz, Wolfgang;Long, Stuart A.;Chen, Ji

文献摘要

参考文献

被引文献

相似文献

PurposeThis本文提出了一种方法来搜索的最坏情况下的配置,导致最高的RF暴露的多配置植入式固定系统在MRI.MethodsA两步方法相结合的人工神经网络和遗传算法的开发,以实现这一目的。在第一步中,使用人工神经网络预测与多配置系统相关的峰值1-g和/或10-g平均比吸收率(SAR 1g/10 g)的RF暴露水平。然后使用遗传算法在枚举离散样本空间和广义连续样本空间内搜索该多维非线性问题的最坏情况配置。作为一个例子,一个通用的板系统,共576配置用于1.5T和3 T MRI systems.ResultsThe所提出的方法可以有效地识别最坏情况下的配置,并准确地预测SAR 1g/10 g与不超过20%的样本在所研究的离散样本空间,甚至可以预测最坏的情况下,在广义连续样本空间。在广义连续样本空间中的最坏情况下的预测误差小于1.6%的SAR 1g和小于1.3%的SAR 10 g相比,与simulation results.ConclusionThe结合的人工神经网络与遗传算法是一种强大的技术来确定最坏情况下的RF暴露水平的多配置系统,只需要少量的训练数据,从整个系统。
PurposeThis paper presents a method to search for the worst‐case configuration leading to the highest RF exposure for a multiconfiguration implantable fixation system under MRI.MethodsA two‐step method combining an artificial neural network and a genetic algorithm is developed to achieve this purpose. In the first step, the level of RF exposure in terms of peak 1‐g and/or 10‐g averaged specific absorption rate (SAR1g/10g), related to the multiconfiguration system, is predicted using an artificial neural network. A genetic algorithm is then used to search for the worst‐case configuration of this multidimensional nonlinear problem within both the enumerated discrete sample space and generalized continuous sample space. As an example, a generic plate system with a total of 576 configurations is used for both 1.5T and 3T MRI systems.ResultsThe presented method can effectively identify the worst‐case configuration and accurately predict the SAR1g/10gwith no more than 20% of the samples in the studied discrete sample space, and can even predict the worst case in the generalized continuous sample space. The worst‐case prediction error in the generalized continuous sample space is less than 1.6% for SAR1gand less than 1.3% for SAR10gcompared with the simulation results.ConclusionThe combination of an artificial neural network with genetic algorithm is a robust technique to determine the worst‐case RF exposure level for a multiconfiguration system, and only needs a small amount of training data from the entire system.
DOI: --
发表时间: 2020
期刊: IEEE transactions on electromagnetic compatibility (Print)
影响因子: --
作者:
Jianfeng Zheng;Xiaohe Ji;W. Kainz;Ji Chen
通讯作者: Ji Chen
使用人工神经网络预测复杂形状医疗植入物的 MRI 射频暴露
DOI: --
发表时间: 2019
期刊: 2019 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting
影响因子: --
作者:
Qianlong Lan;Jianfeng Zheng;Ji Chen
通讯作者: Ji Chen
射频匀场对股骨板植入物附近 MRI 3T 鸟笼线圈引起的局部 SAR 的影响
DOI: --
发表时间: 2017
期刊: 2017 IEEE International Symposium on Antennas and Propagation & USNC/URSI National Radio Science Meeting
影响因子: --
作者:
Qi Zeng;Ran Guo;Jianfeng Zheng;Ji Chen
通讯作者: Ji Chen
多部件骨科髋关节置换系统MRI期间SAR数值研究
DOI: --
发表时间: 2016
期刊: 2016 IEEE International Symposium on Electromagnetic Compatibility (EMC)
影响因子: --
作者:
Jianfeng Zheng;Dawei Li;Ji Chen;W. Kainz
通讯作者: W. Kainz