Alexnet卷积神经网络辨识幽门螺杆菌阳性舌象的可行性研究

目的:探索从舌象图像分析判断幽门螺杆菌(Hp)感染的可行诊断技术。方法:引入Alexnet卷积神经网络概念,通过现代技术对舌象图像分类、识别、计算,以实现对Hp舌象精准客观诊断的功能。结果:通过研究中医舌诊与消化系统的关联、舌象图像可特定性识别的原因以及深度学习Alexnet卷积神经网络结构,论证了该方法对Hp阳性舌象分类模型辨识具有可行性。结论:运用Alexnet卷积神经网络实现辨识Hp感染,将有助于补充与完善中医舌象现代诊断,是实现中医诊疗标准化与客观化的重要技术手段之一。

Feasibility Study on Identification of Helicobacter Pylori Positive Tongue Image by Using Alexnet Convolution Neural Network

Objective:To explore the feasible diagnosis technology of Helicobacter pylori(Hp)infection from tongue picture analysis.Methods:The concept of Alexnet convolution neural network is introduced to classify,recognize and calculate the tongue picture by modern technology,so as to realize the accurate and objective diagnosis of Hp tongue image.Results:By studying the relationship between tongue diagnosis and digestive system in traditional Chinese medicine(TCM),the reason why tongue image can be recognized in a specific way and the structure of Alexnet convolutional neural network in deep learning,the feasibility of this method in Hp positive tongue image classification model is demonstrated.Conclusions:The identification of Hp by using Alexnet convolution neural network will be helpful to supplement and improve the modern diagnosis of tongue image in TCM,and is one of the important technical means to realize the standardization and objectification of TCM diagnosis and treatment.