论文部分内容阅读
目的:评价人工神经网络辅助诊断模型应用于肝癌CT影像诊断的真实性和可靠性。方法:以临床各期肝癌和其它肝占位疾病的CT影像特征数据为训练样本集,建立肝癌CT影像人工神经网络辅助诊断模型,并确定该神经网络应用于肝癌CT影像诊断的截断点;同时,将另外收集的肝癌和其它肝占位疾病CT影像特征数据通过神经网络进行仿真运算,得到网络诊断输出,计算该神经网络模型应用于肝癌CT影像诊断的真实性和可靠性。结果:神经网络用于肝癌CT影像诊断的灵敏度和特异度分别为:98.0%和96.0%,诊断准确度为97.0%;进一步计算该神经网络模型诊断可靠性,其诊断符合率为91.0%,卡帕值为0.82(Z=4.8,P<0.05)。结论:人工神经网络模型可以应用于肝癌CT影像的临床诊断。
Objective: To evaluate the authenticity and reliability of the artificial neural network-assisted diagnostic model in the diagnosis of liver cancer. Methods: CT image features of liver cancer and other liver space occupying diseases in clinical stage were selected as the training samples, and the diagnostic imaging model of liver cancer with CT images was established. The neural network was used to diagnose the CT images in liver cancer. Meanwhile, , The other collection of liver cancer and other liver space occupying disease CT image characteristics of the data through the neural network simulation operation to obtain the network diagnostic output to calculate the neural network model for liver cancer diagnosis of the authenticity and reliability of CT. Results: The sensitivity and specificity of neural network in the diagnosis of liver cancer were 98.0% and 96.0%, respectively, and the diagnostic accuracy was 97.0%. The diagnostic reliability of the neural network model was further evaluated with 91.0% diagnostic accuracy and 91.0% The pawvalue was 0.82 (Z = 4.8, P <0.05). Conclusion: Artificial neural network model can be applied to the clinical diagnosis of liver cancer CT images.