神经网络算法对新冠肺炎CT的判别的准确率研究

Research on the Accuracy of Neural Network Algorithm for the Discrimination of New Coronary Pneumonia CT

随着计算机技术的发展,人工智能在医疗领域的作用日益凸显,尤其是近几年新型的智能影响识别机器人辅助医生进行疾病诊断,大大提升医生诊断效率。本文主要研究神经网络模型对于CT影响的图像识别准确率,以新冠肺炎患者的肺部CT影像为研究对象,进行特征提取,建立卷积神经网络模型和AlexNet深度神经网络模型以及SE-ResNet神经网络模型,最终得到卷积神经网络算法准确率稳定在80%,AlexNet深度神经网络模型的准确率在83%,SE-ResNet神经网络模型的准确率在87%,助力医生快速进行疾病诊断。

With the development of computer technology, the role of artificial intelligence in the medical field has become increasingly prominent. Especially in recent years, the new type of intelligent impact recognition robot assists doctors in disease diagnosis, which greatly improves the efficiency of doc-tors’ diagnosis. This paper mainly studies the image recognition accuracy of the neural network model affected for CT. This article takes the lung CT images of patients with new coronary pneumo-nia as the research object. Feature extraction from CT images. Build the convolutional neural net-work model, AlexNet deep neural network model and SE-ResNet neural network model. Finally, the accuracy of the convolutional neural network algorithm is stable at 80%. The accuracy rate of the AlexNet deep neural network model is 83%, the accuracy rate of SE-ResNet neural network model is 87%, these models can help doctors to quickly diagnose diseases.