基于深度学习的医疗设备智能监控系统设计

Design of intelligent monitoring system for medical equipment based on deep learning

近来,随之医疗行业的发展,医疗设备的安全性问题也逐渐凸显。医疗设备经历长期使用后,可能产生各类运行故障。此次研究联合粒子群算法(Particle Swarm Optimization,PSO)和自适应权重(Adaptive weight)对深度学习的前馈神经网络(Back Propagation,BP)进行了改进,得到APSO-BP算法,并基于此算法设计了医疗设备的智能监控系统。该系统在实验中的测试结果显示,其网络的收敛速度较快,故障识别的均方误差为0.0008324,灵敏度为0.96,特异度为0.92,准确率为0.95,在0.42的风险阈值和0.75的敏感系数下具有0.90的预警准确率。该系统可以对医疗设备的运行状态进行监测,智能识别设备的运行故障并进行风险预警,为医疗设备的周期巡检和维护维修提供有效保障。

Recently,with the development of the medical industry,the safety of medical equipment has gradually become prominent.After long-term use of medical equipment,various operating faults may occur.In this study,Particle Swarm Optimization(PSO)and Adaptive weight improved the deep learning feedforward neural network(Back Propagation(BP),and APSO-BP algorithm was obtained.Based on this algorithm,the intelligent monitoring system of medical equipment is designed.The test results of the system in the experiment show that the convergence speed of the network is fast,the mean square error of fault identification is 0.0008324,the sensitivity is 0.96,the specificity is 0.92,the accuracy is 0.95,and the warning accuracy is 0.90 under the risk threshold of 0.42 and the sensitivity coefficient of 0.75.The system can monitor the running state of medical equipment,intelligently identify the operating faults of equipment and give risk early warning,providing effective guarantee for the periodic inspection and maintenance of medical equipment.