ABSTRACT
Heart sound is one of the common medical signals for diagnosing cardiovascular diseases. This paper studies the binary classification between normal or abnormal heart sounds, and proposes a heart sound classification algorithm based on the joint decision of extreme gradient boosting (XGBoost) and deep neural network, achieving a further improvement in feature extraction and model accuracy. First, the preprocessed heart sound recordings are segmented into four status, and five categories of features are extracted from the signals based on segmentation. The first four categories of features are sieved through recursive feature elimination, which is used as the input of the XGBoost classifier. The last category is the Mel-frequency cepstral coefficient (MFCC), which is used as the input of long short-term memory network (LSTM). Considering the imbalance of the data set, these two classifiers are both improved with weights. Finally, the heterogeneous integrated decision method is adopted to obtain the prediction. The algorithm was applied to the open heart sound database of the PhysioNet Computing in Cardiology(CINC) Challenge in 2016 on the PhysioNet website, to test the sensitivity, specificity, modified accuracy and F score. The results were 93%, 89.4%, 91.2% and 91.3% respectively. Compared with the results of machine learning, convolutional neural networks (CNN) and other methods used by other researchers, the accuracy and sensibility have been obviously improved, which proves that the method in this paper could effectively improve the accuracy of heart sound signal classification, and has great potential in the clinical auxiliary diagnosis application of some cardiovascular diseases.
Subject(s)
Algorithms , Databases, Factual , Heart Sounds , Neural Networks, ComputerABSTRACT
Objective To assess the clinical efficacy and safety of the combination of intracoronary tirofiban infusion(ICTI) plus percutaneous coronary intervention(PCI) in patients with acute ST-elevation myocardial infarction (STEAMI). Methods The 128 cases with STEAMI were enrolled in this study. They were randomly divided into trial group and control group. The 10 μg/kg tirofiban were infused into the infarct related artery (IRA) within 5 minutes through the cather after coronary angiography in trial group (n=64). Normal saline in matched dose was infused into IRA after coronary angiography in control group (n=64). The coronary thrombosis and revascularization status were assessed within 10 minutes after injection. Postoperative bleeding complications were observed in all cases. Adverse cardiovascular events and cardiac function were followed up within 1 month in all cases. Results There were 33 cases whose thrombus burden was reduced within 10 minutes after the infusion of tirofiban in trial group, including 26 cases of thrombolysis in myocardial infarction (TIMI) ≥1. There were 6 cases whose thrombus burden was reduced within 10 minutes after the infusion of normal saline in control group, including 3 cases TIMI ≥ 1. The coronary thrombosis and revascularization status were better in trial group rather than in control group (P＜0.01). Adverse cardiovascular events occurred in 5 cases within 1 month, including 2 cases in trial group and 3 cases in control group (P＞0.05). New York heart association functional class and ejection fraction values were better in trial group rather than in control group within 1 month (P＜0.05). Postoperative bleeding complications occurred in 14 cases by TIMI criteria , including severe and mild bleeding in 2 cases in trial group and 1 cases in control group (P＞0.05), but no significant bleeding occurred in 8 cases in trial group and in 3 cases in control group (P＜0.01). Conclusions The combination of intracoronary infusion of tirofiban plus PCI is effective and safe for thrombolysis and revascularization in IRA in patients with STEAMI.