Objective To explore the independent risk factors for hospital infections in tertiary hospitals in Gansu Province, and establish and validate a prediction model. Methods A total of 690 patients hospitalized with hospital infections in Gansu Provincial Hospital between January and December 2021 were selected as the infection group; matched with admission department and age at a 1∶1 ratio, 690 patients who were hospitalized during the same period without hospital infections were selected as the control group. The information including underlying diseases, endoscopic operations, blood transfusion and immunosuppressant use of the two groups were compared, the factors influencing hospital infections in hospitalized patients were analyzed through multiple logistic regression, and the logistic prediction model was established. Eighty percent of the data from Gansu Provincial Hospital were used as the training set of the model, and the remaining 20% were used as the test set for internal validation. Case data from other three hospitals in Gansu Province were used for external validation. Sensitivity, specificity, accuracy, and area under the receiver operating characteristic curve (AUC) were used to evaluate the model effectiveness. Results Multiple logistic regression analysis showed that endoscopic therapeutic manipulation [odds ratio (OR)=3.360, 95% confidence interval (CI) (2.496, 4.523)], indwelling catheter [OR=3.100, 95%CI (2.352, 4.085)], organ transplantation/artifact implantation [OR=3.133, 95%CI (1.780, 5.516)], blood or blood product transfusions [OR=3.412, 95%CI (2.626, 4.434)], glucocorticoids [OR=2.253, 95%CI (1.608, 3.157)], the number of underlying diseases [OR=1.197, 95%CI (1.068, 1.342)], and the number of surgical procedures performed during hospitalization [OR=1.221, 95%CI (1.096, 1.361)] were risk factors for hospital infections. The regression equation of the prediction model was: logit(P)=–2.208+1.212×endoscopic therapeutic operations+1.131×indwelling urinary catheters+1.142×organ transplantation/artifact implantation+1.227×transfusion of blood or blood products+0.812×glucocorticosteroids+0.180×number of underlying diseases+0.200×number of surgical procedures performed during the hospitalization. The internal validation set model had a sensitivity of 72.857%, a specificity of 77.206%, an accuracy of 76.692%, and an AUC value of 0.817. The external validation model had a sensitivity of 63.705%, a specificity of 70.934%, an accuracy of 68.669%, and an AUC value of 0.726. Conclusions Endoscopic treatment operation, indwelling catheter, organ transplantation/artifact implantation, blood or blood product transfusion, glucocorticoid, number of underlying diseases, and number of surgical cases during hospitalization are influencing factors of hospital infections. The model can effectively predict the occurrence of hospital infections and guide the clinic to take preventive measures to reduce the occurrence of hospital infections.
Objective To construct and verify the diagnostic model of preoperative malignant risk of ovarian tumors, so as to improve the diagnostic efficiency of existing test indexes. Methods The related serological indicators and clinical data of patients with ovarian tumors confirmed by pathology who were treated in the Affiliated Hospital of Southwest Medical University between January 2019 and September 2023 were retrospectively collected, and the patients were randomly divided into a training set and a verification set at a 7∶3 ratio. Logistic regression was used to construct a diagnostic model in the training set, and the diagnostic efficacy of the model was verified through discrimination, calibration, clinical benefit, and clinical applicability evaluation. Results A total of 929 patients with ovarian tumors were included, including 318 cases of malignant ovarian tumors and 611 cases of benign ovarian tumors. The patients were randomly divided into a training set of 658 cases and a validation set of 271 cases. A diagnostic model was constructed using logistic regression in the training set, containing 5 factors namely age, percentage of neutrophil (NEU%), fibrinogen to albumin ratio (FAR), carbohydrate antigen 125 (CA125), and human epididymis protein 4 (HE4): modelUAM=?3.211+0.667×age+2.966×CA125+0.792×FAR+1.637×HE4+0.533×NEU%, with a Hosmer-Lemeshow test P-value of 0.21. The area under the receiver operating characteristic (ROC) curve measured in the training set was 0.927 [95% confidence interval (0.903, 0.951)], the sensitivity was 0.947, and the specificity was 0.780. The area under the ROC curve of the validation set was 0.888 [95% confidence interval (0.840, 0.930)], the sensitivity was 0.744, and the specificity was 0.901. Conclusion A new quantitative tool based on age, NEU%, FAR, CA125 and HE4 can be used for the clinical diagnosis of ovarian malignant tumors, and it is helpful to improve the diagnostic efficiency and is worth popularizing.
Features and interaction between features of liver disease is of great significance for the classification of liver disease. Based on least absolute shrinkage and selection operator (LASSO) and interaction LASSO, the generalized interaction LASSO model is proposed in this paper for liver disease classification and compared with other methods. Firstly, the generalized interaction logistic classification model was constructed and the LASSO penalty constraints were added to the interactive model parameters. Then the model parameters were solved by an efficient alternating directions method of multipliers (ADMM) algorithm. The solutions of model parameters were sparse. Finally, the test samples were fed to the model and the classification results were obtained by the largest statistical probability. The experimental results of liver disorder dataset and India liver dataset obtained by the proposed methods showed that the coefficients of interaction features of the model were not zero, indicating that interaction features were contributive to classification. The accuracy of the generalized interaction LASSO method is better than that of the interaction LASSO method, and it is also better than that of traditional pattern recognition methods. The generalized interaction LASSO method can also be popularized to other disease classification areas.
ObjectiveTo investigate the association of preoperative serum uric acid (UA) levels with postoperative prolonged mechanical ventilation (PMV) in patients undergoing mechanical heart valve replacement.MethodsClinical data of 311 patients undergoing mechanical heart valve replacement in The First Affiliated Hospital of Anhui Medical University from January 2017 to December 2017 were retrospectively analyzed. There were 164 males at age of 55.6±11.4 years and 147 females at age of 54.2±9.8 years. The patients were divided into a PMV group (>48 h) and a control group according to whether the duration of PMV was longer than 48 hours. Spearman's rank correlation coefficient and logistic regression analysis were conducted to evaluate the relationship between preoperative UA and postoperative PMV. The predictive value of UA for PMV was undertaken using the receiver operating characteristic (ROC) curve..ResultsAmong 311 patients, 38 (12.2%) developed postoperative PMV. Preoperative serum UA level mean values were 6.11±1.94 mg/dl, while the mean UA concentration in the PMV group was significantly higher than that in the control group (7.48±2.24 mg/dl vs. 5.92±1.82 mg/dl, P<0.001). Rank correlation analysis showed that UA was positively correlated with postoperative PMV (rs=0.205, P<0.001). Multivariate logistic regression analysis demonstrated that preoperative elevated UA was associated independently with postoperative PMV with odds ratio (OR)=1.44 and confidence interval (CI) 1.15–1.81 (P=0.002). The area under the ROC curve of UA predicting PMV was 0.72, 95% CI0.635–0.806, 6.40 mg/dl was the optimal cut-off value, and the sensitivity and specificity was 76.3% and 63.0% at this time, respectively.ConclusionPreoperative elevated serum UA is an independent risk factor for postoperative PMV in patients undergoing mechanical heart valve replacement and has a good predictive value.
Objective To investigate the effect factors of knee function recovery after operation in distal femoral fractures. Methods From January 2001 to May 2007, 92 cases of distal femoral fracture were treated. There were 50 males and 42 females, aged 20-77 years old (average 46.7 years old). Fracture was caused by traffic accident in 48 cases, by fall ing fromheight in 26 cases, by bruise in 12 cases and by tumble in 6 cases. According to Müller’s Fracture classification, there were 29 cases of type A, 12 cases of type B and 51 cases of type C. According to American Society of Anesthesiologists (ASA) classification, there were 21 cases of grade I, 39 cases of grade II, 24 cases of grade III, and 8 cases of grade IV. The time from injury to operation was 4 hours to 24 days with an average of 7 days. Anatomical plate was used in 43 cases, retrograde interlocking intramedullary nail in 37 cases, and bone screws, bolts and internal fixation with Kirschner pins in 12 cases. After operation, the HSS knee function score was used to evaluate efficacy. Ten related factors were appl ied for statistical analysis, to knee function recovery after operation in distal femoral fractures, such as age, sex, preoperative ASA classification, injury to surgery time, fracture type, treatment, reduction qual ity, functional exercise after operation, whether or not CPM functional training and postoperative compl ications. Results Wound healed by first intention in 88 cases, infection occurred in 4 cases. All patients followed up 16-32 months with an average of 23.1 months. Cl inical union of fracture was achieved within 3-7 months after operation. Extensor device adhesions and the scope of activities of lt; 80° occurred in 29 cases, traumatic arthritis in 25 cases, postoperative fracture displacement in 6 cases, mild knee varus or valgus in 7 cases and implant loosening in 6 cases. According to HSS knee function score, the results were excellent in 52 cases, good in 15 cases, fair in 10 cases and poor in 15 cases with an excellent andgood rate of 72.83%. Single factor analysis showed that age, preoperative ASA classification, fracture type, reduction qual ity, whether or not CPM functional exercise, and postoperative compl ications were significantly in knee function recovery (P lt; 0.05). logistic regression analysis showed that the fracture type, qual ity of reduction, whether or not CPM functional exercise, and age were major factors in the knee joint function recovery. Conclusion Age, preoperative ASA classification, fracture type, reduction qual ity, and whether or not CPM functional training, postoperative compl ications factors may affect the knee joint function recovery. Adjustment to the patient’s preoperative physical status, fractures anatomic reduction and firm fixation, early postoperative active and passive functional exercises, less postoperative compl ications can maximize the restoration of knee joint function.
Objective To investigate the adverse pregnant outcomes of hospitalized pregnant women in Lanzhou city, and analyze the corresponding risk factors and provide basis for the further research on better child-bearing and child-rearing. Methods In two provincial-level hospitals and one provincial-level specialized hospital, the method of cluster random sampling was applied to extract 6 825 medical records from January 2004 to December 2005. The relevant information was abstracted and correlative analyses were undertaken. Results The incidence of adverse pregnancy outcomes for the hospitalized pregnant women in Lanzhou city was 14.65%. Single-factor unconditional logistic regression analyses displayed that the variables with statistical significance were the number of previous pregnancies, the number of previous child delivery, abortion history, abnormal gestation history, and past medical history, whereas multi-factor unconditional logistic regression analyses revealed that the adverse pregnancy outcomes were positively correlated with abnormal gestation history and the number of previous pregnancies with statistical significance. Conclusion The incidence of adverse pregnancy outcomes for the hospitalized pregnant women in Lanzhou city is quite high. Abnormal gestation history and the number of previous pregnancies are the main risk factors for the adverse pregnancy outcomes.
Signal classification is a key of brain-computer interface (BCI). In this paper, we present a new method for classifying the electroencephalogram (EEG) signals of which the features are heterogeneous. This method is called wrapped elastic net feature selection and classification. Firstly, we used the joint application of time-domain statistic, power spectral density (PSD), common spatial pattern (CSP) and autoregressive (AR) model to extract high-dimensional fused features of the preprocessed EEG signals. Then we used the wrapped method for feature selection. We fitted the logistic regression model penalized with elastic net on the training data, and obtained the parameter estimation by coordinate descent method. Then we selected best feature subset by using 10-fold cross-validation. Finally, we classified the test sample using the trained model. Data used in the experiment were the EEG data from international BCI Competition Ⅳ. The results showed that the method proposed was suitable for fused feature selection with high-dimension. For identifying EEG signals, it is more effective and faster, and can single out a more relevant subset to obtain a relatively simple model. The average test accuracy reached 81.78%.
Objective To evaluate the risk factors for postoperative in-hospital mortality in elderly patients receiving cardiac valvular surgery, and develop a new prediction models using the least absolute shrinkage and selection operator (LASSO)-logistic regression. Methods The patients≥65 years who underwent cardiac valvular surgery from 2016 to 2018 were collected from the Chinese Cardiac Surgery Registry (CCSR). The patients who received the surgery from January 2016 to June 2018 were allocated to a training set, and the patients who received the surgery from July to December 2018 were allocated to a testing set. The risk factors for postoperative mortality were analyzed and a LASSO-logistic regression prediction model was developed and compared with the EuroSCOREⅡ. Results A total of 7 163 patients were collected in this study, including 3 939 males and 3 224 females, with a mean age of 69.8±4.5 years. There were 5 774 patients in the training set and 1389 patients in the testing set. Overall, the in-hospital mortality was 4.0% (290/7163). The final LASSO-logistic regression model included 7 risk factors: age, preoperative left ventricular ejection fraction, combined coronary artery bypass grafting, creatinine clearance rate, cardiopulmonary bypass time, New York Heart Association cardiac classification. LASSO-logistic regression had a satisfying discrimination and calibration in both training [area under the curve (AUC)=0.785, 0.627] and testing cohorts (AUC=0.739, 0.642), which was superior to EuroSCOREⅡ. Conclusion The mortality rate for elderly patients undergoing cardiac valvular surgery is relatively high. LASSO-logistic regression model can predict the risk of in-hospital mortality in elderly patients receiving cardiac valvular surgery.
Abstract: Objective To investigate the method of improving effect, by investigating and analyzing the possible risk factors affecting shortterm outcome after total correction of tetralogy of Fallot (TOF). Methods Data of 219 patients who received total correction of TOF were divided into two groups according to the length of postoperative stay in hospital and recovery of heart function in the near future. Group A(n=110): patients had good recovery of heart function classified as gradeⅠorⅡ（NYHA classification）, and could smoothly be discharged from the hospital within two weeks without serious complications. The left ventricular ejection fraction (LVEF) had to exceed to 0.50 during 6 months followup visit. Group B(n=109): patients had worse recovery of heart function classified as grade Ⅱ or Ⅲ, and could not be discharged within two weeks with severe complications. LVEF was less than 0.50 during 6 months followup visit. The clinical data of two groups were compared, and risk factors affecting shortterm outcome after total correction of TOF operation were analyzed by logistic regression and model selection. Results There were good recovery of heart function classified as gradeⅠorⅡ(NYHA classification）in discharge, no death, and LVEF all exceeded to 0.50 in group A; there were 8 deaths in group B (7.34 %), and recovery of heart function was worse classified as grade Ⅱ or Ⅲ, with LVEF being less than 0.50（Plt;0.01）. Amount of postoperative daily thoracic drainage, assisted respiration time, time of inotropic agent stabilizing circulation, and the average length of postoperative stay in group A were all less or short than those in group B（Plt;0.01）. But the bypass and clamping time of group B were exceeded group A. The ratio of patching astride annulus in group B was greater than that in group A, and Nakata index was less than that in group A(Plt;0.01). The results of logistic regression and model selection indicate: age at repair (OR=0.69), oxygen saturation(OR=0.98), haematocrit before operation (OR=0.94), and patching astride annulus (OR=46.86), Nakata index (OR=16.90), amount of postoperative daily thoracic drainage (OR=0.84), presence of arrhythmia(OR=0.87), and wound infection(OR=63.57) have significant effect with shortterm outcome after total correction of TOF operation. Conclusions The probable methods to improving effect of shortterm outcome after total correction of TOF are an earlier age at repair, decreasing haematocrit, rising oxygen saturation before surgery, performing a palliative operation facilitating development of arteriae pulmonalis in earlier time, improving the surgical technique, and strengthening the perioperative care.
ObjectiveTo examine statistical performance of different rare-event meta-analyses methods.MethodsUsing Monte-Carlo simulation, we set a variety of scenarios to evaluate the performance of various rare-event meta-analysis methods. The performance measures included absolute percentage error, root mean square error and interval coverage.ResultsAcross different scenarios, the absolute percentage error and root mean square error were similar for Bayesian logistic regression model, generalized mixed linear effects model and continuity correction, but the interval coverage was higher with Bayesian logistic regression model. The statistical performances with Mantel-Haenszel method and Peto method were consistently suboptimal across different scenarios.ConclusionsBayesian logistic regression model may be recommended as a preferred approach for rare-event meta-analysis.