首页|胸外科护理扫描仪成像故障检测方法研究

胸外科护理扫描仪成像故障检测方法研究

扫码查看
针对传统胸外科护理扫描仪成像故障检测存在检测准确率低和稳定性差的问题,提出一种基于SVR优化深度确定性梯度算法的胸外科护理扫描仪成像故障检测方法.首先,对深度确定性策略梯度算法原理进行具体分析;然后在深度确定性策略梯度算法中引入支持向量机SVR,通过该技术减小估计梯度方差以实现快速收敛;最后将SVR优化深度确定性梯度算法应用到胸外科护理扫描仪成像中,搭建一个故障检测模型,通过此模型实现扫描仪故障准确检测.实验结果表明,本模型对扫描仪成像中偏差故障、偏移故障、完全失效故障和精度下降故障的检测准确率分别取值为 95.42%、96.17%、94.09%和 98.33%,稳定在 90%及以上,且均高于传统的DDPG模型、EMD-MDT模型和KPCA-SVM模型.由此说明,本模型能够实现胸外科护理扫描仪成像故障的准确检测,检测准确率和稳定性显著提升,满足实际应用需求.
Study on imaging fault detection method of nursing scanner in thoracic surgery
Aiming at the problems of low detection accuracy and poor stability,a thoracic surgical nursing scanner imaging fault detection method based on SVR optimized depth certainty gradient algorithm is proposed.Firstly,the principle of depth deterministic strategy gradient algorithm is analyzed;then,support vector machine SVR is introduced into the depth deterministic strategy gradient algorithm to reduce the estimated gradient variance to achieve rapid convergence;finally,SVR optimization depth deterministic gradi-ent algorithm is applied to thoracic nursing scanner imaging to build a fault detection model to realize the accurate fault detection of the scanner.The experimental results show that the detection accuracy of deviation,offset,complete failure and precision failure is 95.42%,96.17%,94.09%and 98.33%respectively,stable at 90%or more,and all higher than the traditional DDPG model,EMD-MDT model and KPCA-SVM model.This shows that this model can realize the accurate detection of imaging faults of thoracic surgery nursing scanner,significantly improve the detection accuracy and stability,and meet the needs of practical application.

thoracic surgical carescanner imagingfault detectiondepth deterministic strategy gradientSVR

邵雯、李纬捷、欧洋

展开 >

西安交通大学第一附属医院,西安 710000

西安交通大学第二附属医院,西安 710000

胸外科护理 扫描仪成像 故障检测 深度确定性策略梯度 SVR

2024

自动化与仪器仪表
重庆工业自动化仪表研究所,重庆市自动化与仪器仪表学会

自动化与仪器仪表

CSTPCD
影响因子:0.327
ISSN:1001-9227
年,卷(期):2024.(11)