Electrocardiogram classification method based on convolutional neural network

Abstract

The invention relates to the technical field of electrocardiosignal processing, in particular to an electrocardiosignal classification method based on a convolutional neural network. The method comprises the steps: S1, collecting electrocardiosignal training data, and attaching labels to the electrocardiosignal training data for data preprocessing; s2, performing data enhancement on the preprocessed training data; s3, constructing a convolutional neural network model, and training the convolutional neural network model by using the enhanced training data to obtain a training model; s4, acquiring a target electrocardiosignal, inputting the target electrocardiosignal into the training model for calculation, and outputting a probability value; and S5, performing positive and negative examplejudgment according to the output probability value to obtain a classification judgment result. According to the method, the electrocardiogram diagnosis efficiency and accuracy can be effectively improved, the provided training model can cover various complex electrocardiogram characteristics, and transfer learning of data is facilitated.

Description

technical field [0001] The invention relates to the technical field of electrocardiogram signal processing, in particular to a method for classifying electrocardiogram based on a convolutional neural network. Background technique [0002] Electrocardiogram examination is a routine inspection item in hospitals. Electrocardiogram is the most basic basis for doctors to judge a patient's heart condition. Electrocardiogram signals are electrical signals converted from non-stationary periodic biological signals caused by heart activity, and contain a large amount of complex heart activity information. At present, there are still some problems in the field of ECG diagnosis: 1. The training cycle of diagnostic doctors is long and the cost is high, and there is still a large gap for high-end talents; 2. The individual subjectivity of doctors varies greatly, and it is difficult to achieve a completely unified diagnostic standard ;3. Doctors only observe with the naked eye, a lot of un...

AI Technical Summary

Problems solved by technology

At present, there are still some problems in the field of ECG diagnosis: 1. The training cycle of diagnostic doctors is long and the cost is high, and there is still a large gap for high-end talents; 2. The individual subjectivity of doctors varies greatly, and it is difficult to achieve a completely unified diagnostic standard ;3. Doctors only observe with the naked eye, a lot of underlying information is invisible, and the information utilization rate is low; 4. There are too many similar diseases, and there is a lot of repetitive labor in the process of viewing pictures; 5. It takes a long time for doctors to read pictures, at least a few minutes to several hours

[0003] Using the traditional digital signal processing method, for a specific disease ECG, first manually extract features, and then make a judgment based on the threshold. This method requires a lot of medical and signal processing experience, and it is not universal. After changing the disease, Difficult to have a higher accuracy rate, or even work at all

The use of machine learning models or statistical learning models has achieved automation to a certain extent and has a certain degree of versatility. Compared with the first method, the accuracy rate has also been greatly improved, but the model still has insufficient expressive ability. problems, cannot cover a variety of complex situations, and is difficult to perform transfer learning

Images