used CNNC for the recognition of five heartbeats and inputted the heartbeat data after segmentation their recognition accuracy reached 94.04% and the model performed well. Compared with experts, the AUC value of their model reached 0.97, and the sensitivity was 0.83, which was higher than the expert level of 0.78.Īnother scholar used CNN to identify the heartbeat and the input-segmented heartbeat, then used the knowledge of CNN to obtain features, and then used more perceptrons to recognize and analyze the input features, which is an end-to-end learning network structure. Hannun A Y et al.’s machine learning group used a 34-layer convolutional neural network to detect arrhythmias. After reviewing the data, we found that the ECG recognition models studied and adopted by many researchers have some limitations regarding the recognition of ECG abnormalities in different populations. The standard database used by most researchers has obvious features and complete data, but in practical use, there are differences in ECG images detected by ECG machines with different leads or different parameters. With the development of technology, the real-time monitoring of dynamic ECGs based on machine learning has also developed rapidly, and the real-time state transmission and abnormality warning of dynamic ECGs promote more accurate and timely diagnoses. Additionally, the convolutional layer with overlapping repetitions in the front segment and pooling layers can perform deep extraction of image features and have better recognition and classification effects on images. The convolutional neural network is a kind of deep neural network commonly used in image recognition, which contains multiple hidden layers while introducing computational operations such as convolution and pooling in the hidden layers to reduce the dimensionality of the image, reduce the parameters of the network, and improve the training efficiency of the neural network, which is now widely used in the recognition and classification of various images, such as those from ECGs. At the same time, manual recognition and diagnosis of ECGs by physicians is very time-consuming, and the long duration of interpretation work will produce fatigue, which will affect the physician’s judgment, thereby promoting misdiagnoses and affecting the accuracy. When different doctors make diagnoses, the results are prone to errors. ![]() Reliable automated interpretation of ECG signals is extremely beneficial for clinical routines and patient safety. ![]() However, due to the large number of physical examination personnel and the heavy workload, there are inevitably mistakes made in the recognition and diagnosis of electrocardiograms by doctors. The ECG is an essential and reliable diagnostic tool used in modern medicine. It is used to assist in the diagnosis of cardiac diseases and plays an important role in determining the effects of drugs or electrolyte conditions on the heart and the status of artificial heart pacing. The lectrocardiogram, which detects electrophysiological signals from the heart muscle, is considered a favourable monitoring method. Additionally, the main ECG abnormalities in these coal company workers were sinus bradycardia, non-specific intraventricular conduction delay, myocardial ischemia, and sinus tachycardia. ![]() Conclusions: The convolutional neural network model can accurately identify the main ECG abnormality types of coal workers. The test set accuracies of the sinus bradycardia model, nonspecific intraventricular conduction delay model, myocardial ischemia model, and sinus tachycardia model were 97.66%, 96.49%, 93.62%, and 93.02%, respectively sensitivities were 96.63%, 96.30%, 96.88% and 95.24%, respectively specificities were 98.78%, 96.67%, 86.67%, and 90.90%, respectively Brier scores were 0.03, 0.07, 0.09, and 0.11, respectively Calibration-in-the-large values were 0.026, 0.110, 0.041, and 0.098, respectively. Results: The number of abnormal ECG results was 849, and the rate of abnormal results was 25.02%. We use Python software and convolutional neural network to establish ECG images recognition and classification model.We usecalibration curve, calibration-in-the-large, Brier score, specificity, sensitivity, F1 score, Kappa value, accuracy, and area under the curve (AUC) of ROC to evaluate the performance of the model. Methods: Coal workers who underwent physical examinations at Gequan Mine Hospital and Dongpang Mine Hospital of Hebei Jizhong Energy from July 2020 to September 2020 were selected as the study subjects. Objective: To process and extract electrocardiogram (ECG, ECG, or EKG) features using a convolutional neural network (CNN) to establish an ECG-assisted diagnosis model.
0 Comments
Leave a Reply. |