ABSTRACT
Objective To establish a hybrid deep learning lung sound classification model based on convolutional neural network(CNN)-long short-term memory(LSTM)for electronic auscultation.Methods Wavelet transform was used to extract features from the dataset,transforming lung sound signals into energy entropy,peak value and other features.On this basis,a classification model based on hybrid algorithm incorporating CNN and LSTM neural network was constructed.The features extracted by wavelet transform were input into CNN module to obtain the spatial features of the data,and then the temporal features were detected through LSTM module.The fusion of the two types of features enabled the classification of lung sounds through the model,thereby assisting in the diagnosis of pulmonary diseases.Results The accuracy rate and F1 score of CNN-LSTM hybrid model were significantly higher than those of other single models,reaching 0.948 and 0.950.Conclusion The proposed CNN-LSTM hybrid model demonstrates higher accuracy and more precise classification,showcasing broad potential application value in intelligent auscultation.
ABSTRACT
Objective:To investigate the effects of aerosol therapy with budesonide suspension combined with compound ipratropium bromide on partial pressure of carbon dioxide (PaCO 2) and tumor necrosis factor α (TNF-α) in children with bronchiolitis. Methods:A total of 124 children with bronchiolitis admitted to Gujiao Central Hospital from January 2019 to December 2021 were included in this study. These children were randomly divided into two groups using the coin-tossing method. The control group ( n = 62) was treated with routine symptomatic treatment, and the study group ( n = 62) was treated with aerosol therapy of budesonide suspension combined with compound ipratropium bromide based on routine symptomatic treatment. The time at which clinical symptoms disappear, clinical efficacy, inflammatory reaction, and blood gas index were determined in each group. Results:After treatment, the time at which asthma, cough, pulmonary rales, and fever in the study group were (2.28 ± 0.71) days, (3.30 ± 0.82) days, (5.25 ± 1.03) days, and (19.01 ± 2.65) hours, respectively, which were significantly shorter than (2.71 ± 0.89) days, (3.81 ± 0.98) days, (5.72 ± 1.37) days, and (20.76 ± 3.12) hours in the control group ( t = 2.97, 3.14, 2.15, 3.36, all P < 0.05). Total response rate and PaO 2 in the study group were 91.94% and (83.94 ± 4.02) mmHg, respectively, which were significantly higher than 77.42% and (81.25 ± 5.53) mmHg in the control group ( χ2 = 5.03, t = 3.09, both P < 0.05). Interleukin-18, interleukin-33, TNF-α, and PaCO 2 in the study group were (141.03 ± 34.69) ng/L, (143.87 ± 38.43) ng/L, (75.49 ± 18.43) ng/L, and (41.85 ± 3.31) mmHg, respectively, which were significantly lower than (158.64 ± 47.92) ng/L, (162.75 ± 50.32) ng/L, (83.22 ± 21.75) ng/L, and (43.58 ± 4.46) mmHg in the control group ( t = -2.34, -2.34, -3.23, -2.45, all P < 0.05). Conclusion:Aerosol therapy with budesonide suspension combined with compound ipratropium bromide based on routine symptomatic treatment is more effective on bronchiolitis than routine symptomatic treatment alone. The combined therapy can effectively decrease PaCO 2 and TNF-α levels.
ABSTRACT
Objective To study the shadowing effect when using UWB bio-radar to detect multiple static human targets to solve the problem in multi-target detection.Methods With simulated breathing apparatus as detection targets,the UWB bioradar multi static targets respiration detection experiment was designed,and the influences of distance and angle between targets and its respiratory frequency and amplitude on the shadowing effect were studied.Result The shadowing effect was mainly affected by the relative position of the multiple targets,while the respiratory frequency and amplitude of the target had less influence on it.Conclusion When multi static human targets are detected the shadowing effect does exist,and the effect mainly derives from the block of electromagnetic wave by the front target,while the change of respiratory parameters of the front target has little influence on the effect.
ABSTRACT
Objective To observe the effects of flunarizine on the intracerebral hemorrhage (ICH) of Wistar rats at different time points. Methods An ICH rat model was established by collagenase. The concentrations of intracellular free Ca 2+ at different time points after ICH were determined by the fluorescence detector with an indicator, Fura 2/AM. The effects of flunarizine on the concentrations at different time points were also observed. Results The concentration of intracellular free Ca 2+ began to increase at 0.5 h, increased significantly at 6 h, and peaked at 24 h (about 4 folds as high as that before hemorrhage), but began to decrease at 72 h after acute intracerebral hemorrhage. Flunarizine decreased the concentration of intracellular free Ca 2+ significantly ( P
ABSTRACT
AIM: To test the effect of ethanol precipitation on the purifying technology of chlorogenic acid (CGA). METHODS: Using HPLC, the effect of the extraction of CGA from the crude Flos Lonicerae Japonica powder with three concentrations of ethanol was compared. RESULTS: Ethanol with concentration of 60% was optimal. CGA content was raised from 5.5% to 37.72% by use of ethanol precipitation and extraction in combination. CONCLUSION: The ethanol precipitation can get rid of the impurity in the crude powder of Flos Lonicerae Japonica effectively and the yield of CGA can reach 90.16%.