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.
ObjectiveTo analyze the factors affecting the prognosis of patients with primary tracheal malignancy, and establish a nomogram model for prediction its prognosis.MethodsA total of 557 patients diagnosed with primary tracheal malignancy from 1975 to 2016 in the Surveillance, Epidemiology, and End Results Data were collected. The factors affecting the overall survival rate of primary tracheal malignancy were screened and modeled by univariate and multivariate Cox regression analysis. The nomogram prediction model was performed by R 3.6.2 software. Using the C-index, calibration curves and receiver operating characteristic (ROC) curve to evaluate the consistency and predictive ability of the nomogram prediction model.ResultsThe median survival time of 557 patients with primary tracheal malignancy was 21 months, and overall survival rates of the 1-year, 3-year and 5-year were 59.1%±2.1%, 42.5%±2.1%, and 35.4%±2.2%. Univariate and multivariate Cox regression analysis showed that age, histology, surgery, radiotherapy, tumor size, tumor extension and the range of lymph node involvement were independent risk factors affecting the prognosis of patients with primary tracheal malignancy (P<0.05). Based on the above 7 risk factors to establish the nomogram prediction model, the C-index was 0.775 (95%CI 0.751-0.799). The calibration curve showed that the prediction model established in this study had a good agreement with the actual survival rate of the 1 year, 3 year and 5 years. The area under curve of 1-year, 3-year and 5-year predicting overall survival rates was 0.837, 0.827 and 0.836, which showed that the model had a high predictive power.ConclusionThe nomogram prediction model established in this study has a good predictive ability, high discrimination and accuracy, and high clinical value. It is useful for the screening of high-risk groups and the formulation of personalized diagnosis and treatment plans, and can be used as an evaluation tool for prognostic monitoring of patients with primary tracheal malignancy.
Objective To analyze the value of serum levels of miR-141-3p, miR-130a, miR-29a-3p, and miR-210 in predicting early neurological deterioration (END) in non-traumatic intracerebral hemorrhage. Methods The patients with non-traumatic cerebral hemorrhage who met the selection criteria and were admitted to Chengde Central Hospital between February 2021 and October 2022 were prospectively selected by convenience sampling method. The serum miR-141-3p, miR-130a, miR-29a-3p, and miR-210 levels upon admission and the occurrence of neurological deterioration within 24 h were collected, and the patients were divided into a deterioration group and a non-deterioration group according to whether neurological deterioration occurred. The correlation of serum miR-141-3p, miR-130a, miR-29a-3p, and miR-210 levels with the END of non-traumatic intracerebral hemorrhage and their predictive value to the END of non-traumatic intracerebral hemorrhage were analyzed. Results A total of 235 patient were enrolled. Of the 235 patients, 45 (19.1%) showed neurological deterioration and 190 (80.9%) showed no neurological deterioration. The levels of miR-141-3p and miR-29a-3p in the deteriorating group were significantly lower than those in the non-deteriorating group [(1.11±0.32) vs. (1.76±0.51) ng/mL, P<0.001; (1.19±0.31) vs. (1.71±0.51) ng/mL, P<0.001], and the levels of miR-130a and miR-210 were significantly higher than those in the non-deteriorating group [(5.13±1.11) vs. (3.82±1.03) ng/mL, P<0.001; (3.96±0.76) vs. (2.78±0.50) ng/mL, P<0.001]. Multivariate logistic regression analysis showed that serum miR-141-3p and miR-29a-3p levels were protective factors for the occurrence of END in non-traumatic intracerebral hemorrhage patients [odds ratio (OR)=0.513, 95% confidence interval (CI) (0.330, 0.798), P=0.003; OR=0.582, 95%CI (0.380, 0.893), P=0.013], and serum miR-130a and miR-210 levels were independent risk factors for that [OR=2.046, 95%CI (1.222, 3.426), P=0.007; OR=2.377, 95%CI (1.219, 4.638), P=0.011]. The area under the receiver operating characteristic curve was 0.857 [95%CI (0.760, 0.954)] in predicting the END of non-traumatic intracerebral hemorrhage by the combined probability of the serum miR-141-3p, miR-130a, miR-29a-3p, and miR-210 levels obtained by logistic regression, and the sensitivity was 86.7%, the specificity was 94.7%, the positive predictive value was 79.6%, and the negative predictive value was 96.8% according to the cut-off value of the prediction probability of the combined test. Conclusion The combined detection of serum miR-141-3p, miR-130a, miR-29a-3p, and miR-210 has a high predictive value in the occurrence of END in non-traumatic intracerebral hemorrhage patients.
ObjectiveTo analyze the risk factors for esophagogastric anastomotic leakage (EGAL) after esophageal cancer surgery, and to establish a risk prediction model for early prevention and treatment.MethodsClinical data of patients undergoing esophagectomy in our hospital from January 2013 to October 2020 were retrospectively analyzed. The independent risk factors for postoperative EGAL were analyzed by univariate and multivariate logistic regression analyses, and a clinical nomogram prediction model was established. According to whether EGAL occurred after operation, the patients were divided into an anastomotic fistula group and a non-anastomotic fistula group.ResultsA total of 303 patiens were enrolled, including 267 males and 36 females with a mean age of 62.30±7.36 years. The incidence rate of postoperative EGAL was 15.2% (46/303). The multivariate logistic regression analysis showed that high blood pressure, chronic bronchitis, peptic ulcer, operation way, the number of lymph node dissected, anastomotic way, the number of intraoperative chest drainage tube, tumor location, no-supplementing albumin in the first three days after operation, postoperative pulmonary infection, postoperative use of bronchoscope were the independent risk factors for EGAL after esophageal cancer surgery (P<0.05). A prognostic nomogram model was established based on these factors with the area under the receiver operating characteristic curve of 0.954 (95%CI 0.924-0.975), indicating a high predictive value.ConclusionThe clinical prediction model based on 11 perioperative risk factors in the study has a good evaluation efficacy and can promote the early detection, diagnosis and treatment of EGAL.
Objective To evaluate the predictive effect of three machine learning methods, namely support vector machine (SVM), K-nearest neighbor (KNN) and decision tree, on the daily number of new patients with ischemic stroke in Chengdu. Methods The numbers of daily new ischemic stroke patients from January 1st, 2019 to March 28th, 2021 were extracted from the Third People’s Hospital of Chengdu. The weather and meteorological data and air quality data of Chengdu came from China Weather Network in the same period. Correlation analyses, multinominal logistic regression, and principal component analysis were used to explore the influencing factors for the level of daily number of new ischemic stroke patients in this hospital. Then, using R 4.1.2 software, the data were randomly divided in a ratio of 7∶3 (70% into train set and 30% into validation set), and were respectively used to train and certify the three machine learning methods, SVM, KNN and decision tree, and logistic regression model was used as the benchmark model. F1 score, the area under the receiver operating characteristic curve (AUC) and accuracy of each model were calculated. The data dividing, training and validation were repeated for three times, and the average F1 scores, AUCs and accuracies of the three times were used to compare the prediction effects of the four models. Results According to the accuracies from high to low, the prediction effects of the four models were ranked as SVM (88.9%), logistic regression model (87.5%), decision tree (85.9%), and KNN (85.1%); according to the F1 scores, the models were ranked as SVM (66.9%), KNN (62.7%), decision tree (59.1%), and logistic regression model (57.7%); according to the AUCs, the order from high to low was SVM (88.5%), logistic regression model (87.7%), KNN (84.7%), and decision tree (71.5%). Conclusion The prediction result of SVM is better than the traditional logistic regression model and the other two machine learning models.
ObjectiveTo explore the CT imaging features and independent risk factors for cystic pulmonary nodules and establish a malignant probability prediction model. Methods The patients with cystic pulmonary nodules admitted to the Department of Thoracic Surgery of the First People's Hospital of Neijiang from January 2017 to February 2022 were retrospectively enrolled. They were divided into a malignant group and a benign group according to the pathological results. The clinical data and preoperative chest CT imaging features of the two groups were collected, and the independent risk factors for malignant cystic pulmonary nodules were screened out by logistic regression analysis, so as to establish a prediction model for benign and malignant cystic pulmonary nodules. ResultsA total of 107 patients were enrolled. There were 76 patients in the malignant group, including 36 males and 40 females, with an average age of 59.65±11.74 years. There were 31 patients in the benign group, including 16 males and 15 females, with an average age of 58.96±13.91 years. Multivariate logistic analysis showed that the special CT imaging features such as cystic wall nodules [OR=3.538, 95%CI (1.231, 10.164), P=0.019], short burrs [OR=4.106, 95%CI (1.454, 11.598), P=0.008], cystic wall morphology [OR=6.978, 95%CI (2.374, 20.505), P<0.001], and the number of cysts [OR=4.179, 95%CI (1.438, 12.146), P=0.009] were independent risk factors for cystic lung cancer. A prediction model was established: P=ex/(1+ex), X=–2.453+1.264×cystic wall nodules+1.412×short burrs+1.943×cystic wall morphology+1.430×the number of cysts. The area under the receiver operating charateristic curve was 0.830, the sensitivity was 82.9%, and the specificity was 74.2%. ConclusionCystic wall nodules, short burrs, cystic wall morphology, and the number of cysts are the independent risk factors for cystic lung cancer, and the established prediction model can be used as a screening method for cystic pulmonary nodules.
Objective To investigate the correlation between monocyte-lymphocyte ratio (MLR) and intensive care unit (ICU) results in ICU hospitalized patients. Methods Clinical data were extracted from Medical Information Mart for Intensive Care Ⅲ database, which contained health data of more than 50000 patients. The main result was 30-day mortality, and the secondary result was 90-day mortality. The Cox proportional hazards model was used to reveal the association between MLR and ICU results. Multivariable analyses were used to control for confounders. Results A total of 7295 ICU patients were included. For the 30-day mortality, the hazard ratio (HR) and 95% confidence interval (CI) of the second (0.23≤MLR<0.47) and the third (MLR≥0.47) groups were 1.28 (1.01, 1.61) and 2.70 (2.20, 3.31), respectively, compared to the first group (MLR<0.23). The HR and 95%CI of the third group were still significant after being adjusted by the two different models [2.26 (1.84, 2.77), adjusted by model 1; 2.05 (1.67, 2.52), adjusted by model 2]. A similar trend was observed in the 90-day mortality. Patients with a history of coronary and stroke of the third group had a significant higher 30-day mortality risk [HR and 95%CI were 3.28 (1.99, 5.40) and 3.20 (1.56, 6.56), respectively]. Conclusion MLR is a promising clinical biomarker, which has certain predictive value for the 30-day and 90-day mortality of patients in ICU.
To screen new tuberculosis diagnostic antigens and vaccine candidates, we predicted the epitopes of Mycobacterium tuberculosis latent infection-associated protein Rv2004c by means of bioinformatics. The homology between Rv2004c protein and human protein sequences was analyzed with BLAST method. The second structures, hydrophilicity, antigenicity, flexibility and surface probability of the protein were analyzed to predict B cell epitopes and T cell epitopes by Protean software of DNAStar software package. The Th epitopes were predicted by RANKPEP and SYFPEITHI supermotif method, the CTL epitopes were predicted by means of combination analyses of SYFPEITHI supermotif method, BIMAS quantitative motif method and NetCTL prediction method. The peptide sequences with higher scores were chosen as the candidate epitopes. Blast analysis showed that Rv2004c protein had low homology with human protein. This protein had abundant secondary structures through analysis of DNAStar software, the peptide segments with high index of hydrophilicity, antigenicity, surface probability and flexibility were widely distributed and were consistent with segments having beta turn or irregular coil. Ten candidates of B cell epitopes were predicted. The Th epitopes of Rv2004c protein were located after the 200th amino acid. Of 37 Th cell epitopes predicted, there were more epitopes of HLA-DRB1*0401 and HLA-DRB1*0701 phenotypes, and the MHC restrictive types of some Th cell epitopes exist cross overlap. Of 10 CTL epitopes predicted, there were more number and higher score of HLA-A2 restricted epitopes. Therefore Mycobacterium tuberculosis Rv2004c protein is a protein antigen with T cell and B cell epitopes, and is expected to be a new target protein candidate for tuberculosis diagnosis and vaccine.
As a common disease in nervous system, epilepsy is possessed of characteristics of high incidence, suddenness and recurrent seizures. Timely prediction with corresponding rescues and treatments can be regarded as effective countermeasure to epilepsy emergencies, while most accidental injuries can thus be avoided. Currently, how to use electroencephalogram (EEG) signals to predict seizure is becoming a highlight topic in epilepsy researches. In spite of significant progress that made, more efforts are still to be made before clinical applications. This paper reviews past epilepsy studies, including research records and critical technologies. Contributions of machine learning (ML) and deep learning (DL) on seizure predictions have been emphasized. Since feature selection and model generalization limit prediction ratings of conventional ML measures, DL based seizure predictions predominate future epilepsy studies. Consequently, more exploration may be vitally important for promoting clinical applications of epileptic seizure prediction.
The risk prediction of paroxysmal atrial fibrillation (PAF) is a challenge in the field of biomedical engineering. This study integrated the advantages of machine learning feature engineering and end-to-end modeling of deep learning to propose a PAF risk prediction method based on multimodal feature fusion. Additionally, the study utilized four different feature selection methods and Pearson correlation analysis to determine the optimal multimodal feature set, and employed random forest for PAF risk assessment. The proposed method achieved accuracy of (92.3 ± 2.1)% and F1 score of (91.6 ± 2.9)% in a public dataset. In a clinical dataset, it achieved accuracy of (91.4 ± 2.0)% and F1 score of (90.8 ± 2.4)%. The method demonstrates generalization across multi-center datasets and holds promising clinical application prospects.