RESUMEN
OBJECTIVE To construct a risk prediction model for bloodstream infection （BSI） induced by carbapenem-resistant Klebsiella pneumoniae （CRKP）. METHODS Retrospective analysis was conducted for clinical data from 253 patients with BSI induced by K. pneumoniae in the First Hospital of Qinhuangdao from January 2019 to June 2022. Patients admitted from January 2019 to December 2021 were selected as the model group （n＝223）， and patients admitted from January 2022 to June 2022 were selected as the validation group （n＝30）. The model group was divided into the CRKP subgroup （n＝56） and the carbapenem- sensitive K. pneumoniae （CSKP） subgroup （n＝167） based on whether CRKP was detected or not. The univariate and multivariate Logistic analyses were performed on basic information such as gender， age and comorbid underlying diseases in two subgroups of patients； independent risk factors were screened for CRKP-induced BSI， and a risk prediction model was constructed. The established model was verified with patients in the validation group as the target. RESULTS Admissioning to intensive care unit （ICU）， use of immunosuppressants， empirical use of carbapenems and empirical use of antibiotics against Gram-positive coccus were independent risk factors of CRKP-induced BSI （ORs were 3.749， 3.074， 2.909， 9.419， 95%CIs were 1.639-8.572， 1.292- 7.312， 1.180-7.717， 2.877-30.840， P＜0.05）. Based on this， a risk prediction model was established with a P value of 0.365. The AUC of the receiver operating characteristic （ROC） curve of the model was 0.848 [95%CI （0.779， 0.916）， P＜0.001]， and the critical score was 6.5. In the validation group， the overall accuracy of the prediction under the model was 86.67%， and the AUC of ROC curve was 0.926 [95%CI （0.809， 1.000]， P＜0.001]. CONCLUSIONS Admission to ICU， use of immunosuppressants， empirical use of carbapenems and empirical use of antibiotics against Gram-positive coccus are independent risk factors of CRKP- induced BSI. The CRKP-induced BSI risk prediction model based on the above factors has good prediction accuracy.
RESUMEN
Objective To study the factors affecting hospital death in elderly patients with novel coronavirus infec-tion/disease 2019(COVID-19),and to build a risk prediction model.Methods According to the diagnostic criteria of Diagnosis and Treatment Protocol for COVID-19 Infection(Trial 10th Edition).Totally 775 elderly patients(≥60 years old)diagnosed as COVID-19 infection in the emergency department and fever clinic of the First Hospital of Changsha were selected as the research objects.General data and serum biomarkers of patients were collected.After treatment,the patients'data were divided into survival group and hospital death group.Binary Logistic regres-sion was used to screen the independent influencing factors of death,and ROC curve was used to analyze the pre-dictive value of related indicators on hospital death.Results After treatment,712 patients(91.9%)survived and 63 patients(8.3%)died in hospital.Binary Logistic regression analysis showed that:≥90 years old[OR=5.065,95%CI(1.427,17.974)],type 2 diabetes mellitus[OR= 3.757,95%CI(1.649,8.559)],COPD[OR= 5.625,95%CI(2.357,13.421)],monocyte ratio[OR=0.908,95%CI(0.857,0.963)],plasma fibringen[OR=1.376,95%CI(1.053,1.800)]and lactate dehydrogenase[OR=1.005,95%CI(1.001,o1.008)]were independent factors of in-hospital death(P＜0.05).The predictive value of diabetes mellitus+COPD+age+monocyte ratio+plasma fibrinogen+lactate dehydrogenase was proved in hospital death from COVID-19 infected patients:the area under the curve(AUC)was 0.883(95%CI:0.827,0.940,P＜0.001),the critical value≥0.710 suggested the risk of death in hospital,the specificity was 0.851,the sensitivity was 0.857.Conclusions The hospital mortality of the elderly after COVID-19 infection is higher and closely related to type 2 diabetes,COPD,monocyte ratio,plasma fibrinogen and lactate dehydrogenase.
RESUMEN
Hepatic encephalopathy is a clinical syndrome of central nervous system dysfunction caused by liver insufficiency.It severely affects the quality of life of patients and may lead to death.Accurate prediction of the risk of developing hepatic encephalopathy is crucial for early intervention and treatment.In order to identify the risk of hepatic encephalopathy in patients in advance,many studies have been devoted to efforts to develop tools and methods to identify the risk of hepatic encephalopathy as early as possible,so as to develop preventive and early management strategies.Most conventional hepatic encephalopathy risk prediction models currently assess the prob-ability of a patient developing hepatic encephalopathy by analysing factors such as clinical data and biochemical indicators,however,their accuracy,sensitivity and positive predictive value are not high.The application of artificial intelligence to clinical predictive modelling is a very hot and promising area,which can use large amounts of data and complex algorithms to improve the accuracy and efficiency of diagnosis and prognosis.To date,there have been few studies using AI techniques to predict hepatic encephalopathy.Therefore,this paper reviews the research progress of hepatic encephalopathy risk prediction models,and also discusses the prospect of AI application in hepatic encephalopathy risk prediction models.It also points out the challenges and future research directions of AI in HE risk prediction model research in order to promote the development and clinical application of hepatic encephalopathy risk prediction models.
RESUMEN
Objective To construct a risk prediction model of pulmonary involvement based on chest CT and clinical feature in patients with primary Sjogren's syndrome(pSS),and to explore the risk prediction value of the model.Methods A total of 360 pSS patients who had been treated at Handan Hospital of Traditional Chinese Medicine from October 2020 to August 2023 were retrospectively selected as study objects,and were then divided into a modeling group(252 patients)and a verification group(108 patients)according to a ratio of 7∶3.The patients in the modeling group were divided into a control group(201 patients)and an involvement group(51 patients)based on presence or absence of lung involvement.The data on clinical characteristics and features of chest high-resolution CT(HRCT)in the modeling group was collected.Univariate analysis was performed among the groups to determine the relevant factors affecting lung involvement in pSS patients.Binary logistic regression analysis was performed on related factors to screen independent risk factors.A prediction model was established based on the independent risk factors.A verification and value analysis of the column-line prediction model were completed through data collection of the verification group.Results Age,disease course,cough,Raynaud's phenomenon,C-reactive protein(CRP),anti-SSA antibody,and HRCT were the relevant factors affecting lung involvement in pSS patients(all P＜0.05).Further binary logistic regression analysis showed that old age,prolonged disease course,cough and abnormal HRCT imaging were independent risk factors for lung involvement in SS patients(all P＜0.05).A nomogram risk prediction model was constructed based on independent factors.The model verification results indicated that the calibration chart showed better performance in the prediction model.The AUC of the area under the receiver operating characteristic(ROC)curve was 0.993 the modeling group and 0.995 in the validation group.Conclusions The clinical characteristics and the results of chest CT are closely related with lung involvement in patients with pSS.Old age,prolonged disease course,cough,and abnormal HRCT imaging are independent risk factors affecting lung involvement in patients with pSS.The prediction model established on this basis has a higher predictive value for the occurrence of lung involvement in patients receiving after-loading radiotherapy.
RESUMEN
BACKGROUND:Proximal humeral fracture in older adults is one of the three major osteoporotic fractures.Anatomic locking plate fixation is the first choice for most scholars to treat difficult-to-reduce and complex fracture types.However,the probability of reduction failure after the operation is high,which seriously affects patients'quality of life. OBJECTIVE:To investigate the correlation between deltoid tuberosity index and postoperative reduction failure of proximal humeral fractures in the elderly,analyze and filter preoperative independent risk factors for reduction failure of proximal humeral fractures in the elderly,and construct and verify the effectiveness of a clinical prediction model. METHODS:The clinical data of 153 elderly patients with proximal humeral fractures who met the diagnosis and inclusion criteria and received open reduction and locking plate surgery in Foshan Hospital of TCM from June 2012 to June 2021 were collected.The patients were divided into the reduction failure subgroup and the reduction maintenance subgroup.The independent risk factors were selected by multivariate Logistic regression analysis,and the nomogram was constructed by R language.After 1000 times of resampling by Bootstrap method,the Hosmer-Lemeshow goodness of fit correlation test,receiver operating characteristic curve,calibration curve,clinical decision,and influence curve were plotted to evaluate its goodness of fit,discrimination,calibration ability,and clinical application value.Fifty-five elderly patients with proximal humeral fractures from June 2013 to August 2021 were selected as the model's external validation group to evaluate the prediction model's stability and accuracy. RESULTS AND CONCLUSION:(1)Of the 153 patients in the training group,44 patients met reduction failure after internal plate fixation.The prevalence of postoperative reduction failure was 28.8%.Multivariate Logistic regression analysis identified that deltoid tuberosity index[OR=9.782,95%CI(3.798,25.194)],varus displacement[OR=4.209,95%CI(1.472,12.031)],and medial metaphyseal comminution[OR=4.278,95%CI(1.670,10.959)]were independent risk factors for postoperative reduction failure of proximal humeral fractures in older adults(P<0.05).(2)A nomogram based on independent risk factors was then constructed.The Hosmer-Lemeshow test results for the model of the training group showed that χ2=0.812(P=0.976)and area under curve=0.830[95%CI(0.762,0.898)].The calibration plot results showed that the model's predicted risk was in good agreement with the actual risk.The decision and clinical influence curves showed good clinical applicability.(3)In the validation group,the accuracy rate in practical applications was 86%,area under curve=0.902[95%CI(0.819,0.985)].(4)It is concluded that deltoid tuberosity index<1.44,medial metaphyseal comminution,and varus displacement were independent risk factors for reduction failure.(5)The internal and external validation of the risk prediction model demonstrated high discrimination,accuracy,and clinical applicability could be used to individually predict and screen the high-risk population of postoperative reduction failure of proximal humeral fractures in the elderly.The predicted number of patients at high risk is highly matched to the actual number of patients who occur when the model's threshold risk probability is above 65%,and clinicians should use targeted treatment.
RESUMEN
Refeeding syndrome(RFS)has a high incidence among critically ill patients and significantly impacts the re-covery and prognosis of the patients.In this paper,we reviewed the literature on the risk factors and risk prediction models for RFS,finding the risk factors of RFS included patient-related,treatment-related factors and disease-related factors and the risk prediction models encompassed risk stratification model,risk score models and the Logistic regression models.It was concluded from the review that early assessment was crucial to preventing the occurrence of RFS.However,there was still a lack of reliable RFS risk prediction models with good predictive performance.It was found as well that it was crucial for the prevention of RFS to attach importance to nutritional and serological indicators and other factors.It was expected to be a necessity to conduct prospec-tive and multicenter studies to develop a risk prediction model for predicting RFS for ICU patients.Our review provides a refer-ence for early assessment and intervention for critically ill patients with RFS.
RESUMEN
Objective To analyze the influencing factors of hypoglycemia in patients undergoing colonoscopy and to construct a risk prediction model and evaluate the model.Methods A total of 528 patients who underwent colonoscopy were selected by the convenience sampling method from the gastroenterology department of a tertiary A hospital in Qingdao from March 2022 to August 2022.Their general information,laboratory indicators and operation-related data were collected.Multivariate Logistic regression was used to analyze the risk factors of hypoglycemia in patients with colonoscopy for risk prediction model construction,and its prediction effect was evaluated by drawing a nomogram.Results Hypoglycemia occurred in 66 of 528 patients,with an incidence of 12.50％.The risk factors finally in the risk prediction model in Logistic regression were drinking history,long fasting time after operation,polyethylene glycol(PEG)-electrolyte solutions＞3 L,low quality of bowel preparation.The model passed Hosmer-Lemeshow goodness of fit test x2=10.158(P=0.200).The area under the ROC curve was 0.829,while the cut-off was 0.575,with sensitivity of 92.90％and specificity of 64.60％.Conclusion Patients undergoing colonoscopy have a higher risk of hypoglycemia.Patients with a history of drinking,longer fasting after surgery,more than 3 L of PEG-electrolyte solutions,and low quality of bowel preparation were more likely to develop hypoglycemia.The established risk prediction model has a good effect,providing the reference for screening high-risk group of hypoglycemia and taking preventive and protective measures.
RESUMEN
Objective:To construct the risk prediction nomogram model of acute kidney injury (AKI) with R language and traditional statistical methods based on the large sample clinical database, and verify the accuracy of the model.Methods:It was a a retrospective case control study. The patients who met the diagnostic criteria of AKI in Tongji Hospital of Tongji University from January 1 to December 31, 2021 were screened in the clinical database, and the patients with monitored serum creatinine within 48 hours but without AKI were included as the control group. The demographic data, disease history, surgical history, medication history and laboratory test data were collected to screen the risk factors of AKI in clinic.Firstly, based on multivariate logistic regression analysis and forward stepwise logistic regression analysis, the selected risk factors were included to construct the nomogram model. At the same time, cross validation, bootstrap validation and randomly split sample validation were used for internal verification, and clinical data of patients in the sane hospital after one year (January to December, 2022) were collected for external verification. The receiver-operating characteristic curve was used to determine the discrimination of the model, and calibration curve and decision curve analysis were carried out to evaluate the accuracy and clinical net benefit, respectively.Results:A total of 5 671 patients were enrolled in the study, with 1 884 AKI patients (33.2%) and 3 787 non-AKI patients (66.7%). Compared with non-AKI group, age, and proportions of surgical history, renal replacement therapy, hypertension, diabetes, cerebrovascular accident,chronic kidney disease, drug use histories and mortality in AKI group were all higher (all P<0.05). Multivariate logistic regression analysis showed that the independent influencing factors of AKI were surgical history, hypertension, cerebrovascular accident, diabetes, chronic kidney disease, diuretics, nitroglycerin, antidiuretic hormones, body temperature, serum creatinine, C-reactive protein, red blood cells, white blood cells, D-dimer, myoglobin, hemoglobin, blood urea nitrogen, brain natriuretic peptide, aspartate aminotransferase, alanine aminotransferase, triacylglycerol, lactate dehydrogenase, total bilirubin, activated partial thromboplastin time, blood uric acid and potassium ion (all P<0.05). Finally, the predictive factors in the nomogram were determined by forward stepwise logistic regression analysis, including chronic kidney disease, hypertension, myoglobin, serum creatinine and blood urea nitrogen, and the area under the curve of the prediction nomogram model was 0.926 [95% CI 0.918-0.933, P<0.001]. The calibration curve showed that the calibration effect of nomogram was good ( P>0.05). The decision curve showed that when the risk threshold of nomogram model was more than 0.04, the model construction was useful in clinic. In addition, the area under the curve of receiver-operating characteristic curve predicted by nomograph model in external validation set was 0.876 (95% CI 0.865-0.886), which indicated that nomograph model had a high discrimination degree. Conclusion:A nomogram model for predicting the occurrence of AKI is established successfully, which is helpful for clinicians to find high-risk AKI patients early, intervene in time and improve the prognosis.
RESUMEN
@#Objective To construct a risk prediction score model for serious adverse event (SAE) after cardiac catheterization in patients with adult congenital heart disease (ACHD) and pulmonary hypertension (PH) and verify its predictive effect. Methods The patients with PH who underwent cardiac catheterization in Wuhan Asian Heart Hospital Affiliated to Wuhan University of Science and Technology from January 2018 to January 2022 were retrospectively collected. The patients were randomly divided into a model group and a validation group according to the order of admission. The model group was divided into a SAE group and a non-SAE group according to whether SAE occurred after the catheterization. The data of the two groups were compared, and the risk prediction score model was established according to the results of multivariate logistic regression analysis. The discrimination and calibration of the model were evaluated using the area under the receiver operating characteristic (ROC) curve and the Hosmer-Lemeshow test, respectively. Results A total of 758 patients were enrolled, including 240 (31.7%) males and 518 (68.3%) females, with a mean age of 43.1 (18.0-81.0) years. There were 530 patients in the model group (47 patients in the SAE group and 483 patients in the non-SAE group) and 228 patients in the validation group. Univariate analysis showed statistical differences in age, smoking history, valvular disease history, heart failure history, N-terminal pro-B-type natriuretic peptide, and other factors between the SAE and non-SAE groups (P<0.05). Multivariate analysis showed that age≥50 years, history of heart failure, moderate to severe congenital heart disease, moderate to severe PH, cardiac catheterization and treatment, surgical general anesthesia, and N-terminal pro-B-type natriuretic peptide≥126.65 pg/mL were risk factors for SAE after cardiac catheterization for ACHD-PH patients (P<0.05). The risk prediction score model had a total score of 0-139 points and patients who had a score>50 points were high-risk patients. Model validation results showed an area under the ROC curve of 0.937 (95%CI 0.897-0.976). Hosmer-Lemeshow goodness-of-fit test: χ2=3.847, P=0.797. Conclusion Age≥50 years, history of heart failure, moderate to severe congenital heart disease, moderate to severe PH, cardiac catheterization and treatment, general anesthesia for surgery, and N-terminal pro-B-type natriuretic peptide≥126.65 pg/mL were risk factors for SAE after cardiac catheterization for ACHD-PH patients. The risk prediction model based on these factors has a high predictive value and can be applied to the risk assessment of SAE after interventional therapy in ACHD-PH patients to help clinicians perform early intervention.
RESUMEN
OBJECTIVE To construct a risk prediction model for non-compliance with inhaled medication in patients with chronic obstructive pulmonary disease （COPD）. METHODS A retrospective analysis was conducted on 365 COPD patients admitted to the cough and wheeze pharmaceutical care clinic of the First Hospital of Qinhuangdao from October 2021 to October 2023. The patients admitted from October 2021 to June 2023 were selected as the model group （n＝303）， and the patients admitted from July to October 2023 were selected as the validation group （n＝62）. The model group was divided into compliance subgroup （n＝126） and non-compliance subgroup （n＝177）. Univariate analysis combined with multivariate Logistic regression analysis were used to analyze the risk factors for non-compliance with inhaled formulations in patients； the risk prediction model was established through regression analysis， and the accuracy of the model prediction was evaluated based on the validation group of patients. RESULTS Multivariate Logistic regression analysis showed that simultaneous use of 2 inhaled formulations （OR＝3.730， 95%CI 1.996-6.971， P＜0.001）， the number of acute exacerbations within one year ≥2 （OR＝2.509， 95%CI 1.509-4.173， P＜0.001）， smoking （OR＝2.167， 95%CI 1.309-3.588， P＝0.003）， complicated with anxiety/depression （OR＝2.112， 95%CI 1.257-3.499， P＝0.004） and mMRC grading≥2 levels （OR＝1.701， 95%CI 1.014-2.853， P＝0.044） were risk factors for non-compliance with inhaled preparations. Based on this， a risk prediction model was established and the ROC curve was drawn. The areas under the curve of the model group and validation group were 0.836 and 0.928， and the overall accuracy of the model’s prediction was 88.71%. CONCLUSIONS The predictive model based on the simultaneous use of 2 inhaled formulations， the number of acute exacerbations within one year ≥2， smoking， complicated with anxiety/depression， mMRC grading ≥2 levels has certain predictive value for the risk of non-compliance with inhaled formulations for COPD patients.
RESUMEN
OBJECTIVE To construct a risk prediction model for non-compliance with inhaled medication in patients with chronic obstructive pulmonary disease （COPD）. METHODS A retrospective analysis was conducted on 365 COPD patients admitted to the cough and wheeze pharmaceutical care clinic of the First Hospital of Qinhuangdao from October 2021 to October 2023. The patients admitted from October 2021 to June 2023 were selected as the model group （n＝303）， and the patients admitted from July to October 2023 were selected as the validation group （n＝62）. The model group was divided into compliance subgroup （n＝126） and non-compliance subgroup （n＝177）. Univariate analysis combined with multivariate Logistic regression analysis were used to analyze the risk factors for non-compliance with inhaled formulations in patients； the risk prediction model was established through regression analysis， and the accuracy of the model prediction was evaluated based on the validation group of patients. RESULTS Multivariate Logistic regression analysis showed that simultaneous use of 2 inhaled formulations （OR＝3.730， 95%CI 1.996-6.971， P＜0.001）， the number of acute exacerbations within one year ≥2 （OR＝2.509， 95%CI 1.509-4.173， P＜0.001）， smoking （OR＝2.167， 95%CI 1.309-3.588， P＝0.003）， complicated with anxiety/depression （OR＝2.112， 95%CI 1.257-3.499， P＝0.004） and mMRC grading≥2 levels （OR＝1.701， 95%CI 1.014-2.853， P＝0.044） were risk factors for non-compliance with inhaled preparations. Based on this， a risk prediction model was established and the ROC curve was drawn. The areas under the curve of the model group and validation group were 0.836 and 0.928， and the overall accuracy of the model’s prediction was 88.71%. CONCLUSIONS The predictive model based on the simultaneous use of 2 inhaled formulations， the number of acute exacerbations within one year ≥2， smoking， complicated with anxiety/depression， mMRC grading ≥2 levels has certain predictive value for the risk of non-compliance with inhaled formulations for COPD patients.
RESUMEN
To establish a disease risk prediction model based on genetic susceptibility genes and environmental risk factors， which can target high-risk population as early as possible， and intervene in the environmental risk factors in this population. Moreover， accurate screening of genetically susceptible populations can enhance the efficiency of health system. In recent years， with the maturation and cost reduction of high-throughput gene testing， gene testing has been widely used in individual clinical decision-making and will play a more important role in medical and health decision-making. The correlation between genetic testing and disease risk prediction is increasing， making it a prominent research topic in this field. This review summarizes the approaches for establishing and evaluating risk prediction models and discusses potential future challenges and opportunities.
RESUMEN
【Objective】 To investigate the effects of preoperative lipid metabolism level on the postoperative prognosis of non-muscular invasive bladder cancer (NMIBC). 【Methods】 Clinical data of NMIBC patients who underwent surgical treatment in our hospital during Mar.2014 and May 2021 were retrospectively analyzed. Based on receiver operating characteristic (ROC) curve, the optimal cutoff values of all lipid metabolism indicators were determined and patients were classified accordingly. The independent risk factors for postoperative recurrence were identified with Cox regression model. The survival was analyzed with Kaplan-Meier, and recurrence-free survival (RFS) was compared using log-rank tests. A recurrence risk prediction model was established based on the high-density lipoprotein (HDL) and other clinic pathological factors and the accuracy of prediction was evaluated with the area under the ROC curve (AUC). 【Results】 Cox multivariate analysis showed HDL, tumor number, tumor size and histological grade were independent risk factors for recurrence (P<0.05). Kaplan-Meier analysis showed that RFS was significantly longer in the high-HDL group than in the low-HDL group (P<0.001). Incorporating HDL, tumor number, tumor size, histological grade, and tumor stage into the recurrence risk model, the AUC was 0.706, and internal cross validation showed the AUC was 0.711. 【Conclusion】 Preoperative HDL is an independent risk factor affecting the RFS of patients with NMIBC, and combining it with clinic pathological factors will improve the prediction of tumor recurrence.
RESUMEN
OBJECTIVE@#To develop and validate a three-year risk prediction model for new-onset cardiovascular diseases (CVD) among female patients with breast cancer.@*METHODS@#Based on the data from Inner Mongolia Regional Healthcare Information Platform, female breast cancer patients over 18 years old who had received anti-tumor treatments were included. The candidate predictors were selected by Lasso regression after being included according to the results of the multivariate Fine & Gray model. Cox proportional hazard model, Logistic regression model, Fine & Gray model, random forest model, and XGBoost model were trained on the training set, and the model performance was evaluated on the testing set. The discrimination was evaluated by the area under the curve (AUC) of the receiver operator characteristic curve (ROC), and the calibration was evaluated by the calibration curve.@*RESULTS@#A total of 19 325 breast cancer patients were identified, with an average age of (52.76±10.44) years. The median follow-up was 1.18 [interquartile range (IQR): 2.71] years. In the study, 7 856 patients (40.65%) developed CVD within 3 years after the diagnosis of breast cancer. The final selected variables included age at diagnosis of breast cancer, gross domestic product (GDP) of residence, tumor stage, history of hypertension, ischemic heart disease, and cerebrovascular disease, type of surgery, type of chemotherapy and radiotherapy. In terms of model discrimination, when not considering survival time, the AUC of the XGBoost model was significantly higher than that of the random forest model [0.660 (95%CI: 0.644-0.675) vs. 0.608 (95%CI: 0.591-0.624), P < 0.001] and Logistic regression model [0.609 (95%CI: 0.593-0.625), P < 0.001]. The Logistic regression model and the XGBoost model showed better calibration. When considering survival time, Cox proportional hazard model and Fine & Gray model showed no significant difference for AUC [0.600 (95%CI: 0.584-0.616) vs. 0.615 (95%CI: 0.599-0.631), P=0.188], but Fine & Gray model showed better calibration.@*CONCLUSION@#It is feasible to develop a risk prediction model for new-onset CVD of breast cancer based on regional medical data in China. When not considering survival time, the XGBoost model and the Logistic regression model both showed better performance; Fine & Gray model showed better performance in consideration of survival time.
Asunto(s)
Humanos , Femenino , Adulto , Persona de Mediana Edad , Adolescente , Neoplasias de la Mama/epidemiología , Enfermedades Cardiovasculares/etiología , Modelos de Riesgos Proporcionales , Modelos Logísticos , China/epidemiologíaRESUMEN
Objective To explore the epidemiological characteristics of pulmonary infection in elderly patients with chronic obstructive pulmonary disease (COPD), and to construct a risk prediction model. Methods Among of 125 elderly patients with COPD from May 2020 to June 2022 were selected as the research subjects. The epidemiological characteristics of infected patients were counted, and the risk factors of pulmonary infection in patients were analyzed and a prediction model was constructed. Results A total of the 125 elderly patients with COPD, there were 46 cases of pulmonary infection, with the infection rate of 36.80%. The detection rate of Gram-negative bacteria was higher than that of Gram-positive bacteria or fungi (64.44% vs 33.33% or 2.22%, P2=0.812 and P=0.295. ROC curve analysis revealed that the AUC value of the prediction model on predicting the pulmonary infection in elderly patients with COPD was 0.802. Conclusion The pathogenic bacteria of elderly patients with COPD complicated with pulmonary infection are mainly Gram-negative bacteria. The prediction model constructed according to the risk factors of pulmonary infection in patients has predictive value on pulmonary infection in patients.
RESUMEN
BACKGROUND@#Previous cardiovascular risk prediction models in Japan have utilized prospective cohort studies with concise data. As the health information including health check-up records and administrative claims becomes digitalized and publicly available, application of large datasets based on such real-world data can achieve prediction accuracy and support social implementation of cardiovascular disease risk prediction models in preventive and clinical practice. In this study, classical regression and machine learning methods were explored to develop ischemic heart disease (IHD) and stroke prognostic models using real-world data.@*METHODS@#IQVIA Japan Claims Database was searched to include 691,160 individuals (predominantly corporate employees and their families working in secondary and tertiary industries) with at least one annual health check-up record during the identification period (April 2013-December 2018). The primary outcome of the study was the first recorded IHD or stroke event. Predictors were annual health check-up records at the index year-month, comprising demographic characteristics, laboratory tests, and questionnaire features. Four prediction models (Cox, Elnet-Cox, XGBoost, and Ensemble) were assessed in the present study to develop a cardiovascular disease risk prediction model for Japan.@*RESULTS@#The analysis cohort consisted of 572,971 invididuals. All prediction models showed similarly good performance. The Harrell's C-index was close to 0.9 for all IHD models, and above 0.7 for stroke models. In IHD models, age, sex, high-density lipoprotein, low-density lipoprotein, cholesterol, and systolic blood pressure had higher importance, while in stroke models systolic blood pressure and age had higher importance.@*CONCLUSION@#Our study analyzed classical regression and machine learning algorithms to develop cardiovascular disease risk prediction models for IHD and stroke in Japan that can be applied to practical use in a large population with predictive accuracy.
Asunto(s)
Humanos , Enfermedades Cardiovasculares/epidemiología , Pronóstico , Estudios Prospectivos , Japón/epidemiología , Accidente Cerebrovascular/etiología , Isquemia Miocárdica/epidemiología , Medición de Riesgo/métodosRESUMEN
Objective To develop a predictive model for postoperative acute kidney injury(AKI)in elderly patients using machine learning methods.Methods The preoperative information and postopera-tive follow-up information of elderly patients who underwent surgery from June 2019 to July 2020 were col-lected,and the laboratory examination results were extracted.A total of 115 preoperative variables were in-cluded.A model of postoperative AKI was constructed using five methods:extreme gradient boosting(XGB),gradient boosting machine(GBM),random forest(RF),support vector machine(SVM),and elastic net logistic regression(ELA).The performance of the model was evaluated using area under the re-ceiver operating characteristic curve(AUROC),area under the precision recall curve(AUPRC),and Brier score.To simplify the model for clinical application,the original model was obtained and some varia-bles with low correlation were removed,and the model was evaluated again using the above method.Results This study ultimately included 5 929 elderly patients,3 359 males(56.7％)and 2 570 females(43.3％),aged 65-99 years.Among them,154 patients(2.6％)experienced postoperative AKI.Among the prediction models constructed using five machine learning methods,XGB has the highest AUROC and AU-PRC,with values of 0.798(95％CI 0.705-0.888)and 0.230(95％CI 0.079-0.374),respectively.Its Brier score is the lowest among all models,the score is 0.023(95％CI 0.014-0.029).After simplifying the XGB model,72 variables were retained.The AUROC of the simplified model was 0.790(95％CI 0.711-0.861),slightly lower than that of the original model.The AUPRC was 0.176(95％CI 0.070-0.313),and the Brier score was 0.024(95％CI 0.017-0.033),and there was no significant statistical difference,indicating that there was no significant difference in the predictive ability of the simplified model compared to the original model.Conclusion Among the five machine learning methods used to construct postoperative AKI prediction models,XGB has the best predictive performance.The simplified XGB predic-tion model still retains high predictive performance and is easier to be promoted in clinical practice.
RESUMEN
Objective Constructing a risk prediction model of IgA nephropathy proteinuria treated by traditional Chinese medicine based on random survival forest model,Screening prognostic risk factors of IgA nephropathy proteinuria.Methods Collecting retrospectively clinical data of 129 cases diagnosed with IgA nephropathy,randomly divided them into training set(60%)and test set(40%).The risk prediction model of IgA nephropathy proteinuria was constructed in the training set with the random survival forest model,and the prognostic risk factors were screened by VIMP method.The accuracy of risk prediction model was validated in the test set with time-dependent ROC curve(tdROC).Results According to the result of VIMP,the prognostic risk factors for IgA nephropathy proteinuria are in the order of eGFR,hypertension,traditional Chinese medicine,24 hUPRO>1 g,genomo sclerosis ratio,Lee grading,fat,hyperlipidemia,hypertrophymia,hyparmane ledmia,Anemia,age and gender.The eGFR was negatively and non-linearly associated with the risk rate of developing persistent proteinuria.Glomerulosclerosis ratio greater than 0.3 is approximately linearly and positively associated with the risk rate of persistent proteinuria.Conclusion Random survival forest model has good predictive performance in the risk prediction model of IgA nephropathy proteinuria treated by traditional Chinese medicine.This risk model can determine the result of IgA nephropathy treated by traditional Chinese medicine,and which is helpful for clinical follow-up monitoring and formulation of individualized treatment plans.
RESUMEN
Objective:To construct and validate a risk prediction model for immune checkpoint inhibitor-associated pneumonia (CIP) using machine learning algorithms and the nomogram, aiming to provide an accurate and intuitive method to assist nurses in screening people at high risk of developing CIP.Methods:This was a retrospective case -control study. A total of 230 oncology patients treated with immune checkpoint inhibitors attending Zhujiang Hospital of Southern Medical University from January 2019 to February 2022 were collected using the hospital's electronic medical record system. The prediction models were built using five machine learning algorithms and nomogram. The models were then validated on a separate test set, and their differentiation and stability were assessed using evaluation indices like AUC and accuracy rate.Results:Underlying lung disease, smoking history, serum albumin≤35 g/L and radiotherapy history were identified as important influencing factors of CIP in all six models. The AUC of K nearest neighbor, support vetor machines (SVM), naive Bayesian, decision tree and random forest models predicted CIP were 0.647, 0.696, 0.930, 0.870, and 0.934, respectively. The AUC of the model created by the nomogram was 0.813, which was lower than the best random forest model in the machine learning algorithm, but with good predictive performance (AUC=0.934).Conclusions:The nomogram model can assess the patient′s risk more intuitively, but the risk prediction model of CIP based on a machine learning algorithm has a higher diagnostic value. It is suggested that the accuracy and usefulness of the prediction model can be increased by combining the nomogram's foundation with the machine learning algorithm.
RESUMEN
Objective:To construct a Bayesian network risk prediction model for delirium during recovery from general anesthesia. To explore the network relationship between awakening delirium of general anesthesia and its related factors, and to reflect the influence intensity of each factor on awakening delirium of general anesthesia through network reasoning.Methods:This is a cross-sectional study. From February to May 2022, the Chinese version of the four rapid delirium diagnosis protocols for general anesthesia patients admitted to the department of Anesthesia, the First Hospital of Shanxi Medical University were adopted as research subjects through convenience sampling method to carry out the delirium screening program during awakening, and general information and blood sample laboratory test results of the subjects were collected. The single factor analysis was used to screen the correlative factors of awakening delirium and a Bayesian network model based on the maximum minimum climb method (MMHC) was constructed.Results:A total of 480 patients were included in the study, and the delirium rate during the recovery period of general anesthesia was 12.9%(62/480). The Bayesian network of awakening delirium consisted of 11 nodes and 18 directed edges. The Bayesian network showed that age, sodium, cerebral infarction and hypoproteinemia were the direct factors related to awakening delirium, while ASA grade, hematocele and hemoglobin were the indirect factors related to awakening delirium. The area under its ROC curve was 0.80(0.78-0.83).Conclusions:Bayesian networks can well reveal the complex network connections between awakening delirium and its related factors, and then prevent and control awakening delirium accordingly.