ABSTRACT
In order to enhance the accuracy of computer aided electrocardiogram analysis, we propose a deep learning model called CBRNN to assist diagnosis on electrocardiogram for clinical medical service. It combines two sub networks which are convolutional neural network (CNN) and bi-directional recurrent neural network (BRNN). In the model, CNN with one-dimension convolution is employed to extract features for each lead of ECG, and BRNN is used to fuse features of different leads to represent deeper features. In the training step, we use more than 40 thousand training data and more than 19 thousand validation data to obtain the optimal parameters of the model. Besides, by validating our model on more than CCDD 120,000 real data, it achieves an 87.69% accuracy rate, higher than popular deep learning models such as CNN and ResNet. Our model has better accuracy than state-of-the-art models and it is also slightly higher than the average accuracy of human judgement. It can be served for the first round screening of ECG examination clinical diagnosis.
Subject(s)
Cardiology , Deep Learning , Diagnosis, Computer-Assisted/methods , Electrocardiography , Signal Processing, Computer-Assisted , Skin/pathology , Algorithms , Humans , Machine Learning , Medical Errors , Medical Informatics , Models, Cardiovascular , Neural Networks, ComputerABSTRACT
This study aims to discuss the operative skills of hysteroscopic tubal embolization and reduce the occurrence of complications.Ninety-four patients were divided into group A and group B. The main surgical technique in group A: when the inner sleeve is sent to the fallopian tube and no longer accessible (but no >3âcm), remove the guide wire and put into the microcoil. But in group B, there are four major surgical techniques. First, the depth at which the guide wire enters the tube was controlled at 2âcm. Second, the inner diameter of the fallopian tube must be explored to determine the type and shape of the coils. Third, saline should be used to separate the catheter. Fourth, it is to control the release speed of the coils. The superiority of the improved operation method was confirmed by comparing the surgical failure rate, incidence of complications, and cost of surgery before and after the procedure.The reoperation rate of group A was 10% (3/30), while that of group B was 2.68% (3/112). The ectopic microcoils rate of group A was 6.67% (2/30), while that of group B was 0.89% (1/112). The microcoil damages rate of group 23.33% (7/30), while that of group B was 8.04% (9/112). All P values were <.01, and the difference was statistically significant.Hysteroscopic tubal embolization is currently a new surgical procedure to block the fallopian tubes and prevent the reverse flow of fluid in the fallopian tubes into the uterine cavity. After we improved surgical techniques, the surgical failure rate, complication rate, and operation cost of fallopian tube embolization were significantly lower than before the improved method was applied. The improved techniques led to a higher success rate.