ABSTRACT
Objective To establish a lung cancer risk prediction model using data mining technology and compare the performance of decision tree C5.0 and artificial neural networks in the application of risk prediction model, and to explore the value of data mining techniques in lung cancer risk prediction. Methods We collected the data of 180 patients with lung cancer and 240 patients with benign lung lesion which contained 17 variables of risk factors and clinical symptoms. Decision tree C5.0 and artificial neural networks models were established to compare the prediction performance. Results There were 420 valid samples collected in total and proportioned with the ratio of 7:3 for the training set and testing set. The accuracy, sensitivity, specificity, Youden index, positive predictive value, negative predictive value and AUC of artificial neural networks model were 65.3%, 61.7%, 73.3%, 0.350, 54.9%, 73.1% and 0.675 (95%CI: 0.628-0.720) in testing set; those of decision tree C5.0 model were 61.0%, 47.8%, 80.4%, 0.282, 35.3%, 80.6% and 0.641 (95%CI: 0.593-0.687) in testing set. Conclusion The artificial neural networks model is superior to the decision tree C5.0 model at overall performance and it has potential application value in the risk prediction of lung cancer.
ABSTRACT
Objective To study the clinical effects of pulmonary recruitment maneuvers combined with pressure regulation volume control (PRVC) in the treatment of severe respiratory distress syndrome (RDS) in premature infants.Method From July 2015 to September 2016,preterm infants of grade Ⅲ-Ⅳ RDS who received PRVC treatment in neonatal department of Huai'an Maternal and Child Health Hospital were assigned into recruitment maneuver group and control group (without recruitment maneuver) using randon number table.The ventilator parameters were observed at 1,2,6,12,18 h and 24 h after ventilation.Recovery rate,duration of oxygen therapy and ventilation,duration of hospital stay,incidence of second dose of pulmonary surfactant and complications were compared between two groups.Result A total of 18 cases were included in recruitment maneuver group and 19 cases in control group.The recovery rate of recruitment maneuver group was higher than control group (16/18 vs.10/19).The duration of oxygen therapy [(6.6 ± 2.3) d vs.(11.8 ± 3.0) d],duration of ventilation [(4.1 ± 2.3) d vs.(6.4 ± 2.8) d],duration of hospital stay [(26.7 ± 7.0) d vs.(33.0 ± 8.4) d] in recruitment maneuver group were significantly shorter than control group (P ＜ 0.05).The proportion of bronchopulmonary dysplasia (1/18 vs.8/19),retinopathy of premature (1/18 vs.7/19),patent ductus arteriosus that require medication closure (1/18 vs.7/19)and incidence of second dose of pulmonary surfactant (2/18 vs.9/19) in recruitment maneuver group were significantly lower than control group (P ＜ 0.05).While the complication of air leak,necrotizing enteritis,Ⅲ-V grade intracranial hemorrhage showed no significant differences between the two groups (P ＞ 0.05).Conclusion Recruitment maneuvers combined with PRVC in treatment of severe RDS premature infants can improve recovery rate and oxygenation.It can also shorten the duration of oxygen therapy,ventilation and hospital stay.It can reduce the incidence of bronchopulmonary dysplasia and retinopathy of premature.It is worth spreading in clinical practice.reduce the incidence of bronchopuhmonary dysplasia and retinopathy.It is worthy of promotion.
ABSTRACT
Objective:To compare the effectiveness between three methods for purifying the immunoglobulin of egg yolk(IgY) which are polyethylene glycol (PEG) method, chloroform extraction method and chloroform / PEG method, and to provide basis for obtaining the batch of IgY.Methods:The inactivated vaccine of Vibrio parahemolyticus (V. parahemolyticus) was prepared and the hens were immunized by multi-point intramuscular injection.The eggs were collected and the IgY was purified by PEG method, chloroform extraction method and chloroform/PEG method.The protein extraction rate, the IgY titer and the purity of the antibody which purified by different methods were detected.Furthermore, the operation process, cost and safety of the three methods were analyzed.Results:The protein contents of the extraction belonging three methods from high to low in turn were chloroform extraction method, chloroform/PEG method, and PEG method.There was no significant difference in the antibody titer between three methods, and the tiler of chloroform extraction method was slightly high.The purities of purified antibody from high to low in turn were PEG method, chloroform/PEG method and chloroform method.The PEG method had better security but relatively lower extraction efficiency and higher cost.The chloroform/PEG method had high extraction efficiency and good antibody purity.Conclusion:The PEG method is suitable for a small amount of extraction in the laboratory.The chloroform/PEG method is appropriate for extracting the high quality IgY in a batch as it has high extraction efficiency and good antibody purity.
ABSTRACT
Objective To explore the clinical significance of Downes score in predicting respiratory failure (RF) in term neonates.Methods From June 1, 2011 to December 31, 2013, 243 hospitalized term neonates with dyspnea, who were born at 37 to 42 gestational age and admitted to the hospital within 24 h after birth, were selected from Neonatal Medical Center of Huai'an Maternity and Child Healthcare.Downes scores were evaluated based on five indicators including respiratory rate, skin color, three depressions sign, moaning and lung breath sounds at the time of admission.RF was diagnosed based on clinical menifestations and radial arterial blood gas analysis.Receiver operating characteristic (ROC) curve analysis was performed to evaluate the predictive value of Downes score on RF.Results The average Downes score of the 243 cases was 6.57±0.86 and the RF rate was 69.6％ (169/243).The RF rate was 93.4％ (169/181) when Downes score ≥ 4.5, with the sensitivity, specificity, misdiagnostic rate and missed diagnositic rate of 100.0％, 83.8％, 16.2％ and 0.0％, respectively.While, the RF rate was 80.7％ (130/161) when Downes score ≥ 5.5, with the sensitivity, specificity, misdiagnostic rate and missed diagnositic rate of 76.9％, 58.1％, 41.9％ and 23.1％, respectively.When Downes score ≥ 6.5, the incidence of RF was 97.6％ (123/126), with the sensitivity, specificity, misdiagnositic rate and missed diagnositic rate of 72.8％, 95.9％, 4.1％ and 27.2％, respectively.The cutoff value was 6.5 according to the Youden Index, which with lower rate of misdiagnosis and missed diagnosis.When Downes score ≥ 7.5, the incidence of RF was 100.0％(22/22), with the sensitivity, specificity, misdiagnositic rate and missed diagnosis rate of 13.1％, 100.0％, 0.0％ and 87.0％, respectively.The area under the ROC curve was 0.906 (95％CI: 0.866-0.955).The higher the Downes score, the greater the possibility of RF (P=0.000).Conclusions Dynamic Downes scores in term neonates with dyspnea could predict the risk of RF.The higher the Downes score, the greater the possibility of RF.