OBJECTIVES: This study investigates independent risk factors for postoperative internal fixation device infection in patients with maxillofacial fractures and proposes an early warning model based on the synthetic minority over-sampling technique (SMOTE) algorithm. METHODS: A total of 1 104 patients who underwent surgical treatment for maxillofacial fractures at Oral and Maxillofacial Surgery Department, Affiliated Hospital of Nantong University from January 2021 to December 2024 were retrospectively analyzed. The patients were divided into two groups based on the presence of postoperative internal fixation device infection: the infection group (27 cases) and non-infection group (1 077 cases). Clinical data from both groups were collected and subjected to statistical analysis. Univariate and binary Logistic regression analysis were used to identify risk factors for postoperative internal fixation device infection in maxillofacial fractures. Subsequently, a Logistic regression model was established, and the dataset was improved based on the SMOTE algorithm to construct an early warning model with the improved dataset. The prediction performance of the models was compared and validated. RESULTS: Among the 1 104 patients who underwent surgical treatment for maxillofacial fractures, 27 cases of postoperative internal fixation device infections were identified, corresponding to an infection rate of 2.45% (27/1 104). Age, diabetes history, fracture severity, and oral hygiene status were all identified as risk factors for postoperative internal fixation device infections in maxillofacial fractures (all P<0.05). The prediction model based on the original data (P1). The prediction model based on the SMOTE algorithm (P2). Receiver operating characteristic (ROC) curve analysis shows that the area under curve (AUC) for the P2 model was 0.882, the P1 model was 0.861, indicating the superior predictive performance of the P2 model. The DeLong test results show that the difference in AUC between the two models was statistically significant (P<0.05). CONCLUSIONS Age, diabetes history, postoperative fracture severity, and oral hygiene status are all risk factors for infections associated with internal fixation devices after maxillofacial fracture surgery. The proposed early warning model demonstrated good predictive performance. Medical professionals can utilize this model to effectively intervene and anticipate infections related to internal fixation devices after maxillofacial fracture surgery.
Humans ; Algorithms ; Retrospective Studies ; Male ; Female ; Fracture Fixation, Internal/instrumentation* ; Risk Factors ; Middle Aged ; Adult ; Logistic Models ; Surgical Wound Infection/epidemiology* ; Aged ; Internal Fixators/adverse effects* ; Maxillofacial Injuries/surgery* ; Adolescent

OBJECTIVES: This study aimed to use machine learning algorithms to build a prediction model of the first permanent molar caries of 9-year-old children in Suzhou and screen out risk factors. METHODS: Random stratified whole group sampling was applied to randomly select 9-year-old students from 38 primary schools in 14 townships and streets in Wuzhong District for oral examination and questionnaire survey. Multifactor Logistics regression was used to analyze the risk factors of tooth decay. The data set was randomly divided into training sets and verification sets according to 8∶2, and R 4.3.1 was used to build five machine learning algorithms: random forest, decision tree, extreme gradient boosting (XGBoost), Logistics regression, and lightweight gradient enhancement (LightGBM). The predictive effect of these five models was evaluated using the area under the characteristic curve (AUC). The marginal contribution of quantitative characteristics to the caries prediction model was determined through Shapley additive explanations (SHAP). RESULTS: This study included 7 225 samples that met the standard. The caries rate of the first permanent molar was 54.96%. Multifactor Logistic regression analysis showed that sweet drinks, dessert and candy, snack frequency, and snacks before going to bed after brushing teeth were correlated with the occurrence of first permanent molar caries (P<0.05). The AUC values of decision tree, Logistic regression, LightGBM, random forest, and XGBoost were 75.5%, 83.9%, 88.6%, 88.9%, and 90.1%, respectively. Compared with the variables after single heat coding, the SHAP value of high-frequency sweets (such as dessert candy ≥2 times a day, mother's sugary diet ≥2 times a day) and bad oral hygiene habits (such as frequent snacks before going to bed after brushing teeth and irregular brushing teeth) exhibited the highest positive. CONCLUSIONS XGBoost algorithm has a good prediction effect for first permanent molar caries in 9-year-old children. High-frequency sweet factors and bad oral hygiene habits have a strong positive impact on the risk of first permanent molar caries and are key drivers that can be used in the formulation of targeted interventions.
Humans ; Dental Caries/epidemiology* ; Child ; Machine Learning ; China/epidemiology* ; Molar ; Risk Factors ; Female ; Logistic Models ; Male ; Decision Trees ; Algorithms

Objective To analyze the risk factors for prolonged length of hospital stay in patients with peritoneal dialysis associated peritonitis(PDAP)and construct a nomogram based on Logistic regression model.Methods A retrospective study was conducted on patients with PDAP who were hospitalized at the Affiliated Hospital of Jiangsu University from January 2013 to December 2023.Using the 75th percentile of hospitalization time as the cutoff(>21 days),the patients were divided into prolonged length of hospital stay group and normal length of hospital stay group.Clinical data were compared between the two groups.Logistic regression analysis was used to analyze the risk factors for prolonged hospital stay in PDAP patients and to construct a nomogram.Results A total of 131 PDAP patients were included in this study,including 40 cases in prolonged length of hospital stay group and 91 cases in normal length of hospital stay group.Multivariate Logistic regression analysis showed that Gram-negative bacteria detected in ascites(OR=6.012,95% CI=1.878-19.248,P=0.003)and elevated platelet count(OR=1.010,95% CI=1.005-1.015,P<0.001)were independent risk factors for prolonged length of hospital stay,while elevated serum chloride(OR=0.885,95% CI=0.802-0.978,P=0.016)was a protective factor.Based on the above three indicators,a nomogram was constructed.The multivariate Logistic regression model showed an area under the receiver operating characteristic curve(AUC)of 0.755,with an internal validation AUC of 0.727 using the Bootstrap method.The calibration curve indicated that the predicted probability was consistent with the actual probability.The decision curve showed that the model was clinically applicable when the threshold probabilities were 9%-10%,13% and 18%-92%.Conclusion A nomogram,based on the detection of gram-negative bacteria in ascites,platelet count and serum chloride,was helpful for clinical screening PADP patients at risk for prolonged length of hospital stay,and can provide a basis for optimizing clinical decision-making.
Humans ; Nomograms ; Risk Factors ; Peritoneal Dialysis/adverse effects* ; Retrospective Studies ; Length of Stay ; Peritonitis/etiology* ; Logistic Models ; Male ; Female ; Middle Aged ; Aged

Objective To analyze the influencing factors of death in the elderly with coronavirus disease 2019(COVID-19).Methods The case data of death caused by COVID-19 in West China Fourth Hospital from January 1 to July 8,2023 were collected,and surviving cases from the West China Elderly Health Cohort infected with COVID-19 during the same period were selected as the control.LASSO-Logistic regression was adopted to analyze the data after propensity score matching and the validity of the model was verified by drawing the receiver operating characteristic curve.Results A total of 3 239 COVID-19 survivors and 142 deaths with COVID-19 were included.The results of LASSO-Logistic regression showed that smoking(OR=3.33,95%CI=1.46-7.59,P=0.004),stroke(OR=3.55,95%CI=1.15-10.30,P=0.022),malignant tumors(OR=19.93, 95%CI=8.52-49.23, P<0.001),coronary heart disease(OR=7.68, 95%CI=3.52-17.07, P<0.001),fever(OR=0.51, 95%CI=0.26-0.96, P=0.042),difficulty breathing or asthma symptoms(OR=21.48, 95%CI=9.44-51.95, P<0.001),and vomiting(OR=8.19,95%CI=2.87-23.58, P<0.001)increased the risk of death with COVID-19.The prediction model constructed based on the influencing factors achieved an area under the curve of 0.889 in the test set.Conclusions Smoking,stroke,malignant tumors,coronary heart disease,fever,breathing difficulty or asthma symptoms,and vomiting were identified as key factors influencing the death risk in COVID-19.
Humans ; COVID-19/mortality* ; Aged ; Propensity Score ; China/epidemiology* ; Risk Factors ; Logistic Models ; Smoking ; SARS-CoV-2 ; Male ; Female ; Stroke ; Neoplasms

Objective To evaluate the diagnostic value of ultrasound in chronic appendicitis. Methods A retrospective analysis was conducted on the ultrasound imaging features of the appendixes in 68 patients with chronic appendicitis (chronic appendicitis group) confirmed by pathological results at the Affiliated Hospital of Inner Mongolia Medical University from January 2023 to December 2024,as well as 85 healthy volunteers (normal appendix group) of different ages with no history of abdominal pain during the same period.Multivariate Logistic regression was employed to investigate the sensitivity and specificity of different variables in diagnosing chronic appendicitis. Results The chronic appendicitis group had higher appendix diameter (Z=-8.47,P<0.001),unilateral wall thickness (Z=-7.16,P<0.001),and submucosal thickness (Z=-9.73,P<0.001) than the normal appendix group.Appendix diameter (OR=3.11,95%CI=1.37-7.02,P=0.006) and submucosal thickness (OR=5 492.73,95%CI=89.53-336 984.13,P<0.001) were identified as independent factors for diagnosing chronic appendicitis,while gender,age,unilateral wall thickness,and intraluminal conditions had no significant impact on the diagnosis of chronic appendicitis (all P>0.05).When appendix diameter combined with submucosal thickness was used as a joint diagnostic indicator,the model demonstrated the best performance,with the sensitivity of 92.65%,the specificity of 97.65%,and the accuracy increasing to 95.42%. Conclusion The combined use of appendix diameter and submucosal thickness can significantly improve the accuracy,specificity,and reliability of ultrasound in diagnosing chronic appendicitis.
Humans ; Appendicitis/diagnostic imaging* ; Ultrasonography ; Retrospective Studies ; Male ; Female ; Adult ; Chronic Disease ; Middle Aged ; Appendix/diagnostic imaging* ; Sensitivity and Specificity ; Young Adult ; Logistic Models ; Adolescent

OBJECTIVE: To analyze the risk factors of acute spinal cord injury complicated with respiratory dysfunction, and to construct the clinical prediction model of acute spinal cord injury complicated with respiratory dysfunction. METHODS: Continuous 170 cases of acute spinal cord injury treated from April 2019 to October 2022 were retrospectively collected, and clinical data were uniformly collected. Patients were divided into respiratory dysfunction group 30 cases and non-respiratory dysfunction group 140 cases according to whether they had respiratory dysfunction during treatment. The predictive factors of acute spinal cord injury complicated with respiratory dysfunction were screened by Lasso analysis, and the risk factors of acute spinal cord injury complicated with respiratory dysfunction were screened by multivariate Logistic regression analysis. R(R4.2.1) software was used to establish a nomogram risk warning model for predicting acute spinal cord injury complicated with respiratory dysfunction, and Hosmer-Lemeshow test was used to evaluate the model fit. Finally, area under receiver operating characteristic(ROC) curve (AUC), calibration curve, and decision curve analysis(DCA) were used to evaluate the differentiation, calibration and clinical impact of the model. RESULTS: The incidence of respiratory dysfunction in 170 patients was 17.65%. Lasso regression analysis selected age, residence, marital status, smoking, hypertension, degree of paralysis, spinal cord injury plane, multiple injuries, spinal cord fracture and dislocation, and ASIA grade as the influencing factors. Multivariate Logistic regression analysis showed that age, smoking, degree of paralysis, level of spinal cord injury, spinal cord injury of fracture and dislocation, and ASIA grade were risk factors for acute spinal cord injury complicated with respiratory dysfunction. The prediction model of acute spinal cord injury complicated with respiratory dysfunction was established by Hosmer-Lemeshow test, χ²=5.830, P=0.67. The AUC value of the model was 0.912. DCA analysis showed that the net benefit value of nomogram prediction of acute spinal cord injury complicated with respiratory dysfunction was higher when threshold probability ranged from 1% to 100%. CONCLUSION This column chart can help identify the risk of acute spinal cord injury complicated with respiratory dysfunction in early clinical stage, facilitate early clinical decision-making and intervention, and has important guiding significance for optimizing clinical efficacy and improving prognosis of patients. It is expected to improve and verify this model with larger samples and multi-center in the future.
Humans ; Spinal Cord Injuries/complications* ; Nomograms ; Male ; Female ; Middle Aged ; Adult ; Retrospective Studies ; Risk Factors ; Aged ; Respiration Disorders/etiology* ; Adolescent ; Logistic Models

Objective To mine and analyze the routine blood test data of children with allergic rhinitis (AR), identify routine blood parameters related to childhood allergic rhinitis, establish an effective diagnostic model, and evaluate the performance of the model. Methods This study was a retrospective study of clinical cases. The experimental group comprised a total of 1110 children diagnosed with AR at the First Affiliated Hospital of Air Force Medical University during the period from December 12, 2020 to December 12, 2021, while the control group included 1109 children without a history of allergic rhinitis or other allergic diseases who underwent routine physical examinations during the same period. Information such as age, sex and routine blood test results was collected for all subjects. The levels of routine blood test indicators were compared between AR children and healthy children using comprehensive intelligent baseline analysis, with indicators of P≥0.05 excluded; variables were screened by Lasso regression. Binary Logistic regression was used to further evaluate the influence of multiple routine blood indexes on the results. Five kinds of machine model algorithms were used, namely extreme value gradient lift (XGBoost), logistic regression (LR), gradient lift decision tree (LGBMC), Random forest (RF) and adaptive lift algorithm (AdaBoost), to establish the diagnostic models. The receiver operating characteristic (ROC) curve was used to screen the optimal model. The best LightGBM algorithm was used to build an online patient risk assessment tool for clinical application. Results Statistically significant differences were observed between the AR group and the control group in the following routine blood test indicators: mean cellular hemoglobin concentration (MCHC), hemoglobin (HGB), absolute value of basophils (BASO), absolute value of eosinophils (EOS), large platelet ratio (P-LCR), mean platelet volume (MPV), platelet distribution width (PDW), platelet count (PLT), absolute values of leukocyte neutrophil (W-LCC), leukocyte monocyte (W-MCC), leukocyte lymphocyte (W-SCC), and age. Lasso regression identified these variables as important predictors, and binary Logistic regression further analyzed the significant influence of these variables on the results. The optimal machine learning algorithm LightGBM was used to establish a multi-index joint detection model. The model showed robust prediction performance in the training set, with AUC values of 0.8512 and 0.8103 in the internal validation set. Conclusion The identified routine blood parameters can be used as potential biomarkers for early diagnosis and risk assessment of AR, which can improve the accuracy and efficiency of diagnosis. The established model provides scientific basis for more accurate diagnostic tools and personalized prevention strategies. Future studies should prospectively validate these findings and explore their applicability in other related diseases.
Humans ; Male ; Female ; Rhinitis, Allergic/blood* ; Child ; Biomarkers/blood* ; Retrospective Studies ; Early Diagnosis ; Child, Preschool ; ROC Curve ; Logistic Models ; Hematologic Tests ; Algorithms ; Adolescent ; Machine Learning

Objective To investigate the causal relationship between serum total bile acid (TBA) levels and gastric cancer (GC) using regression discontinuity design (RDD). Methods A total of 1244 GC patients and 1333 healthy controls were included in the study. Demographic characteristics, gallbladder disease history, tumor markers, and serum TBA levels were collected from both groups. Logistic regression was used to construct a risk prediction model to estimate the risk of GC. RDD was employed with serum TBA as the grouping variable and the individual risk of developing GC as the outcome variable. Results The predictive factors in the GC risk prediction model included age, sex, body mass index(BMI), serum TBA, carcinoembryoniv antigen(CEA), alpha fetoprotein(AFP), carbohydrate antigen 199(CA199), and CA125. Serum TBA was identified as an independent risk factor for GC (OR=1.054, 95% CI: 1.030 to 1.079). RDD analysis indicated that when serum TBA levels reached 8 μmol/L, the probability of developing GC increased sharply by 23.7%. The breakpoint remained statistically significant following validity and robustness assessments. Conclusion The study demonstrates a positive causal relationship between serum TBA levels and GC, when the serum TBA level reaches 8 μmol/L, the risk of an individual developing GC increases sharply.
Humans ; Stomach Neoplasms/etiology* ; Male ; Female ; Middle Aged ; Bile Acids and Salts/blood* ; Aged ; Adult ; Risk Factors ; Case-Control Studies ; Biomarkers, Tumor/blood* ; Logistic Models

PURPOSE: We carried out the study aiming to explore and analyze the risk factors, the distribution of pathogenic bacteria, and their antibiotic-resistance characteristics influencing the occurrence of surgical site infection (SSI), to provide valuable assistance for reducing the incidence of SSI after traumatic fracture surgery. METHODS: A retrospective case-control study enrolling 3978 participants from January 2015 to December 2019 receiving surgical treatment for traumatic fractures was conducted at Tangdu Hospital of Air Force Medical University. Baseline data, demographic characteristics, lifestyles, variables related to surgical treatment, and pathogen culture were harvested and analyzed. Univariate analyses and multivariate logistic regression analyses were used to reveal the independent risk factors of SSI. A bacterial distribution histogram and drug-sensitive heat map were drawn to describe the pathogenic characteristics. RESULTS: Included 3978 patients 138 of them developed SSI with an incidence rate of 3.47% postoperatively. By logistic regression analysis, we found that variables such as gender (males) (odds ratio (OR) = 2.012, 95% confidence interval (CI): 1.235 - 3.278, p = 0.005), diabetes mellitus (OR = 5.848, 95% CI: 3.513 - 9.736, p < 0.001), hypoproteinemia (OR = 3.400, 95% CI: 1.280 - 9.031, p = 0.014), underlying disease (OR = 5.398, 95% CI: 2.343 - 12.438, p < 0.001), hormonotherapy (OR = 11.718, 95% CI: 6.269 - 21.903, p < 0.001), open fracture (OR = 29.377, 95% CI: 9.944 - 86.784, p < 0.001), and intraoperative transfusion (OR = 2.664, 95% CI: 1.572 - 4.515, p < 0.001) were independent risk factors for SSI, while, aged over 59 years (OR = 0.132, 95% CI: 0.059 - 0.296, p < 0.001), prophylactic antibiotics use (OR = 0.082, 95% CI: 0.042 - 0.164, p < 0.001) and vacuum sealing drainage use (OR = 0.036, 95% CI: 0.010 - 0.129, p < 0.001) were protective factors. Pathogens results showed that 301 strains of 38 species of bacteria were harvested, among which 178 (59.1%) strains were Gram-positive bacteria, and 123 (40.9%) strains were Gram-negative bacteria. Staphylococcus aureus (108, 60.7%) and Enterobacter cloacae (38, 30.9%) accounted for the largest proportion. The susceptibility of Gram-positive bacteria to Vancomycin and Linezolid was almost 100%. The susceptibility of Gram-negative bacteria to Imipenem, Amikacin, and Meropenem exceeded 73%. CONCLUSION Orthopedic surgeons need to develop appropriate surgical plans based on the risk factors and protective factors associated with postoperative SSI to reduce its occurrence. Meanwhile, it is recommended to strengthen blood glucose control in the early stage of admission and for surgeons to be cautious and scientific when choosing antibiotic therapy in clinical practice.
Humans ; Surgical Wound Infection/epidemiology* ; Male ; Female ; Risk Factors ; Retrospective Studies ; Middle Aged ; Adult ; Case-Control Studies ; Fractures, Bone/surgery* ; Aged ; Drug Resistance, Bacterial ; Logistic Models ; Anti-Bacterial Agents/therapeutic use* ; Incidence ; Bacteria/drug effects*

PURPOSE: Emergency resuscitative thoracotomy (ERT) is a final salvage procedure for critically injured trauma patients. Given its low success rate and ambiguous indications, its use in blunt trauma scenarios remains highly debated. Consequently, our study seeks to ascertain the overall survival rate of ERT in blunt trauma patients and determine which patients would benefit most from this procedure. METHODS: A retrospective case-control study was conducted for this research. Blunt trauma patients who underwent ERT between January 2020 and December 2023 in our trauma center were selected for analysis, with the endpoint outcome being in-hospital survival, divided into survival and non-survival groups. Inter-group comparisons were conducted using Chi-square and Fisher's exact tests, the Kruskal-Wallis test, Student's t-test, or the Mann-Whitney U test. Univariate and multivariate logistic regression analyses were conducted to assess potential predictors of survival. Then, the efficacy of the predictors was assessed through sensitivity and specificity analysis. RESULTS: A total of 33 patients were included in the study, with 4 survivors (12.12%). Multivariate logistic regression analysis indicated a significant association between cardiac tamponade and survival, with an adjusted odds ratio of 33.4 (95% CI: 1.31 - 850.00, p = 0.034). Additionally, an analysis of sensitivity and specificity, targeting cardiac tamponade as an indicator for survivor identification, showed a sensitivity rate of 75.0% and a specificity rate of 96.6%. CONCLUSION The survival rate among blunt trauma patients undergoing ERT exceeds traditional expectations, suggesting that select individuals with blunt trauma can significantly benefit from the procedure. Notably, those presenting with cardiac tamponade are identified as the subgroup most likely to derive substantial benefits from ERT.
Adult ; Female ; Humans ; Male ; Middle Aged ; Case-Control Studies ; China ; Logistic Models ; Resuscitation/mortality* ; Retrospective Studies ; Survival Rate ; Thoracotomy/methods* ; Trauma Centers/statistics & numerical data* ; Wounds, Nonpenetrating/surgery*