Cardiac auscultation is the basic way for primary diagnosis and screening of congenital heart disease(CHD). A new classification algorithm of CHD based on convolution neural network was proposed for analysis and classification of CHD heart sounds in this work. The algorithm was based on the clinically collected diagnosed CHD heart sound signal. Firstly the heart sound signal preprocessing algorithm was used to extract and organize the Mel Cepstral Coefficient (MFSC) of the heart sound signal in the one-dimensional time domain and turn it into a two-dimensional feature sample. Secondly, 1 000 feature samples were used to train and optimize the convolutional neural network, and the training results with the accuracy of 0.896 and the loss value of 0.25 were obtained by using the Adam optimizer. Finally, 200 samples were tested with convolution neural network, and the results showed that the accuracy was up to 0.895, the sensitivity was 0.910, and the specificity was 0.880. Compared with other algorithms, the proposed algorithm has improved accuracy and specificity. It proves that the proposed method effectively improves the robustness and accuracy of heart sound classification and is expected to be applied to machine-assisted auscultation.
ObjectiveTo compare the clinical efficacy between right visual double lumen tube (VDLT) intubation and right common double lumen tube (DLT) intubation in lung isolation technique. MethodsA total of 57 patients undergoing thoracoscopic surgery with right DLT lung isolation technique in the First People's Hospital of Chenzhou City and West China Hospital from June 2020 to June 2021 were randomly divided into two groups: a DLT group (n=29, 16 males and 13 females, with a mean age of 54.3±13.2 years) and a VDLT group (n=28, 18 males and 10 females, with a mean age of 55.1±13.7 years) at 1 : 1 with random number table generated by the computer. The clinical data of the two groups were compared. ResultsCompared with the DLT group, the catheter positioning time in the VDLT group was statistically shorter (74.9±47.5 s vs. 151.6±88.9 s, P<0.001), the right upper lung occlusion rate (21.4% vs. 51.7%) and the intraoperative re-adjustment catheterization rate (14.3% vs. 48.3%) were lower (P<0.05). The quality of lung collapses immediately after thoracotomy (67.9% vs. 24.1%) and 20 minutes after thoracotomy (100.0% vs. 75.9%) were improved (P<0.05). There was no significant difference in the rate of fiberoptic bronchoscope assistance for positioning, or the incidence of pharynx pain and hoarseness between the two groups (P>0.05). ConclusionCompared with common DLT, VDLT is more efficient, accurate and intuitive in the location of right bronchial intubation.