ABSTRACT
Transarterial chemoembolization (TACE) is an established therapeutic strategy for intermediate stage Barcelona Clinic Liver Cancer (BCLC) hepatocellular carcinoma (HCC). However, patients who are early refractory to TACE may not benefit from repeated TACE treatment. Our primary objective was to assess the diagnostic value of inflammatory markers in identifying early TACE refractory for patients with early (BCLC 0 and A) or intermediate (BCLC B) stage HCC. We retrospectively reviewed the HCC patients who underwent TACE as the initial treatment in two hospitals. Patients with early TACE refractoriness had significantly poorer median overall survival (OS) (16 vs 40 months, P<0.001) and progression-free survival (PFS) (7 vs 23 months, P<0.001) compared to TACE non-refractory patients. In the multivariate regression analysis, tumor size (P<0.001), bilobular invasion (P=0.007), high aspartate aminotransferase-to-platelet ratio index (APRI) (P=0.007), and high alpha fetoprotein (AFP) level (P=0.035) were independent risk factors for early TACE refractoriness. The predictive model showcasing these factors exhibited high ability proficiency, with an area under curve (AUC) of 0.833 (95%CI=0.774-0.892) in the training cohort, 0.750 (95%CI: 0.640-0.861) in the internal-validation cohort, and 0.733 (95%CI: 0.594-0.872) in the external-validation cohort. Calibration curve analysis revealed good agreement between the actual and predicted probabilities of early TACE refractoriness. Our preliminary study estimated the potential value of inflammatory markers in predicting early TACE refractoriness and provides a predictive model to assist in identifying patients who may not benefit from repeat TACE treatment.
ABSTRACT
Abstract We aimed to develop a prognostic model for primary pontine hemorrhage (PPH) patients and validate the predictive value of the model for a good prognosis at 90 days. A total of 254 PPH patients were included for screening of the independent predictors of prognosis, and data were analyzed by univariate and multivariable logistic regression tests. The cases were then divided into training cohort (n=219) and validation cohort (n=35) based on the two centers. A nomogram was developed using independent predictors from the training cohort to predict the 90-day good outcome and was validated from the validation cohort. Glasgow Coma Scale score, normalized pixels (used to describe bleeding volume), and mechanical ventilation were significant predictors of a good outcome of PPH at 90 days in the training cohort (all P<0.05). The U test showed no statistical difference (P=0.892) between the training cohort and the validation cohort, suggesting the model fitted well. The new model showed good discrimination (area under the curve=0.833). The decision curve analysis of the nomogram of the training cohort indicated a great net benefit. The PPH nomogram comprising the Glasgow Coma Scale score, normalized pixels, and mechanical ventilation may facilitate predicting a 90-day good outcome.
ABSTRACT
Objective To investigate the risk factors of postoperative fecal contamination in children pa-tients with Hirschsprung's disease(HSCR),and to construct and evaluate the risk predictive model.Methods The clinical data in 377 children patients with HSCR in 3 class 3A hospitals in Guangxi from Janu-ary 2016 to June 2021were retrospectively analyzed by adopting the convenience sampling method.The pa-tients were divided into the modeling group(n=264)and testing model group(n=113)with a ratio of 7∶3.The risk factors of postoperative fecal soiling were analyzed by the single factor and multiple factors,and the risk predictive model was constructed.The receiver operating characteristic(ROC)curve was used to detect the discriminative ability of the model and the H-L test was used to determine the goodness of fit of the mod-el.The model was prospectively validated in 21 children patients with HSCR from August to December 2021.Results Among 377 children patients with HSCR,the fecal soiling occurred in 131 cases with a incidence rate of 34.75%.The constructed predictive model of fecal contamination risk after HSCR operation:logit(P)=-2.385+1.697 × special type of megacolon+0.929 × Soave+0.105 × length of bowel resection+2.065 × il-literate caregivers+0.808 × caregivers'implementation of postoperative diet+0.867 × postoperative defecation training by caregivers.The area under the curve(AUC)in the modeling group was 0.849,the Yoden index was 0.53,the optimal critical value of the model was 0.32,the sensitivity was 76.00%,and the specificity was 77.00%.The H-L test,X2=6.649,P=0.575.AUC of the testing model group was 0.736,the sensitivity was 81.25%,and the specificity was 78.46%.The prospective validation results showed that the sensitivity and specificity of the model were 66.67%and 100%respectively.Conclusion The constructed model has good i-dentification and predictive ability.
ABSTRACT
Objective To explore the risk factors of complicating urogenic sepsis after percutaneous nephrolithotripsy(PCNL)and construct a nomogram prediction model.Methods The data of 291 patients with stage 1 PCNL in 940 Hospital of Joint Logistics Support Force from October 2016 to October 2021 were retrospectively analyzed.The patients were divided into the sepsis group and non-sepsis group according to whether complicating urogenic sepsis after operation.The general data,stone-related data,operation-related data and laboratory detection related data were included.The independent risk factors were screened by the univariate and multivariate logistic regression analysis,and the nomogram prediction model was constructed.Results The results of univariate and multivariate logistic regression analysis showed that age≥60 years old(OR=6.438,95%CI:1.548-26.769),urinary leukocyte 3+(OR=5.651,95%CI:1.614-31.766),urinary nitrite positive(OR=7.117,95%CI:1.190-42.561),operation time≥90 min(OR=4.626,95%CI:1.137-18.817)and perfusion volume 30 L(OR=3.312,95%CI:1.090-10.061)were the independent risk factors of postoperative complicating urogenic sepsis.C-index of the constructed nomogram prediction model in the modeling samples was 0.937,the calibrated C-index was 0.914,and the model predictive efficien-cy was good.Conclusion Age ≥60 years old,urinary leukocyte 3+,urinary nitrite positive,operation time 90 min and perfusion volume ≥30 L are the independent risk factors for complicating urogenic sepsis after PCNL;the constructed nomogram prediction model has a good predictive efficiency for the occurrence of post-operative urogenic sepsis.
ABSTRACT
Objective:The predictive model of cardiac arrest in the emergency room was constructed and validated based on Logistic regression.Methods:This study was a retrospective cohort study. Patients admitted to the emergency room of the First Affiliated Hospital of Xinjiang Medical University from January 2020 to July 2021 were included. The general information, vital signs, clinical symptoms, and laboratory examination results of the patients were collected, and the outcome was cardiac arrest within 24 hours. The patients were randomly divided into modeling and validation group at a ratio of 7:3. LASSO regression and multivariable logistic regression were used to select predictive factors and construct a prediction model for cardiac arrest in the emergency room. The value of the prediction model was evaluated using the area under the receiver operator characteristic curve (AUC), calibration curve, and decision curve analysis (DCA).Results:A total of 784 emergency room patients were included in the study, 384 patients occurred cardiac arrest. The 10 variables were ultimately selected to construct a risk prediction model for cardiac arrest: Logit( P)= -4.503+2.159×modified early warning score (MEWS score)+2.095×chest pain+1.670×abdominal pain+ 2.021×hematemesis+2.015×cold extremities+5.521×endotracheal intubation+0.388×venous blood lactate-0.100×albumin+0.768×K ++0.001×D-dimer. The AUC of the model group was 0.984 (95% CI: 0.976-0.993) and that of the validation group was 0.972 (95% CI: 0.951-0.993). This prediction model demonstrates good calibration, discrimination, and clinical applicability. Conclusions:Based on the MEWS score, chest pain, abdominal pain, hematemesis, cold extremities, tracheal intubation, venous blood lactate, albumin, K +, and D-dimer, a predictive model for cardiac arrest in the in-hospital emergency room was constructed to predict the probability of cardiac arrest in emergency room patients and adjust the treatment strategy in time.
ABSTRACT
Lung cancer is the malignant tumor with the highest incidence and mortality among the Chinese.Tumor node metastasis(TNM)staging established by the American Joint Committee on Cancer(AJCC)and International Union Against Cancer(UICC)is a commonly used criterion,but it still has limitations in judging the prognosis of non-small cell lung cancer(NSCLC)patients.With the advantages of real-time and convenient sampling,the immune score based on peripheral blood biomarkers have the ability to predict prognosis and efficacy of NSCLC patients,which have been developed and validated in clinical studies.However,clinical impleruentation of peripheral immune scores is still not widely in NSCLC patients.Therefore,this study introduces and evaluates the 6 peripheral immune scores and reviews the reseach progress of them in the treatment of NSCLC.
ABSTRACT
Objective:To construct a Chinese neonatal model of early-onset sepsis (EOS) using the Kaiser Permanente sepsis risk calculator and laboratory indicators and validate its clinical prediction potential.Methods:Newborns with a gestational age of ≥34 weeks, who were hospitalized in the Department of Neonatology, the First Affiliated Hospital of Zhengzhou University from January 2020 to June 2022 were retrospectively recruited.Their clinical data were collected.Predictors were screened via the multivariate regression analysis, and the Nomogram model was constructed using R software and RStudio software.Hosmer-Lemeshow test, receiver operating characteristic curve, the decision curve analysis (DCA) were used to evaluate the prediction potential of the Nomogram.Results:A total of 769 patients were enrolled, including 107 patients in the EOS group (5 culture-confirmed cases and 102 clinically diagnosed cases), and 662 cases in the non-EOS group.Ten variables were screened and introduced into the Nomogram, including the gestational age, birth weight, body temperature, white blood cell count, C-reactive protein, procalcitonin, premature rupture of membranes≥18 h, infection of Group B Streptococcus, ventilator application, and prenatal antibiotics.The predictive model showed good discrimination and consistency, with the area under the curve of 0.834 (95% CI: 0.771-0.896). The DCA of the prediction model showed that it was effective in clinical application within the effective threshold of 6%-95%, with a net benefit following the application of corresponding treatment measures. Conclusions:A Chinese neonatal model of EOS was created by using the Kaiser Permanente sepsis risk calculator and laboratory indicators, which has been validated effective.It provides references for clinical management and the guidance for the use of antibiotics.
ABSTRACT
Objective:To explore the relationship between risk perception and health promoting lifestyle profile in population with cardiovascular disease (CVD), and construct a prediction model for clinical screening and targeted intervention.Methods:A cross-sectional survey method was used to select 272 people at moderate and high risk of CVD from the Second Affiliated Hospital of Zhejiang University School of Medicine from March to August 2022. The general information questionnaire, Chinese version of Attitude and Beliefs about Cardiovascular Disease Knowledge and Risk Questionnaire (ABCD-C), and Health Promoting Lifestyle Profile-II (HPLP Ⅱ) were used. Based on multiple regression analysis, a nomogram model for health promoting lifestyle in high-risk CVD population was constructed.Results:Among 272 participants, male 150 cases, female 122 cases, aged (60.58 ± 10.64) years old. The total ABCD-C score was (56.57 ± 5.69), and the total HPLP Ⅱ score was (111.92 ± 12.47). ABCD-C score was significantly positively correlated with HPLP Ⅱ score ( r=0.556, P<0.01). The median of HPLP Ⅱ total score (111 points) was used as the cut-off point for low level of health-promoting lifestyle (≤111 points) and high level of health-promoting lifestyle (>111 points), and used it as the dependent variable, smoking ( OR=0.215, 95% CI 0.104-0.446) was a barrier factor for participants to adopt healthy lifestyle; being married ( OR=14.237, 95% CI 1.963-103.238), having a family average monthly income higher than 5 000 yuan ( OR=4.101, 95% CI 1.369-12.288), higher score of CVD prevention knowledge ( OR=1.660, 95% CI 1.373-2.007), perceived benefits and intention to change physical activity ( OR=1.445, 95% CI 1.255-1.663), perceived benefits and intention to change healthy diet ( OR=1.322, 95% CI 1.058-1.654) were promoting factors. Conclusions:The health-promoting lifestyle of populations at risk for CVD is above-average, influenced by factors such as smoking, marital and economic status, risk attitudes, and beliefs. Utilizing the nomogram model for early screening and targeted risk communication among key populations may contribute to improving their health behavior.
ABSTRACT
BACKGROUND:Based on different algorithms of machine learning,how to carry out clinical research on lumbar disc herniation with the help of various algorithmic models has become a trend and hot spot in the development of intelligent medicine at present. OBJECTIVE:To review the characteristics of different algorithmic models of machine learning in the diagnosis and treatment of lumbar disc herniation,and summarize the respective advantages and application strategies of algorithmic models for the same purpose. METHODS:The computer searched PubMed,Web of Science,EMBASE,CNKI,WanFang,VIP and China Biomedical(CBM)databases to extract the relevant articles on machine learning in the diagnosis and treatment of lumbar disc herniation.Finally,96 articles were included for analysis. RESULTS AND CONCLUSION:(1)Different algorithm models of machine learning provide intelligent and accurate application strategies for clinical diagnosis and treatment of lumbar disc herniation.(2)Traditional statistical methods and decision trees in supervised learning are simple and efficient in exploring risk factors and establishing diagnostic and prognostic models.Support vector machine is suitable for small data sets with high-dimensional features.As a nonlinear classifier,it can be applied to the recognition,segmentation and classification of normal or degenerative intervertebral discs,and to establish diagnostic and prognostic models.Ensemble learning can make up for the shortcomings of a single model.It has the ability to deal with high-dimensional data and improve the precision and accuracy of clinical prediction models.Artificial neural network improves the learning ability of the model,and can be applied to intervertebral disc recognition,classification and making clinical prediction models.On the basis of the above uses,deep learning can also optimize images and assist surgical operations.It is the most widely used model with the best performance in the diagnosis and treatment of lumbar disc herniation.The clustering algorithm in unsupervised learning is mainly used for disc segmentation and classification of different herniated segments.However,the clinical application of semi-supervised learning is relatively less.(3)At present,machine learning has certain clinical advantages in the identification and segmentation of lumbar intervertebral discs,classification and grading of the degenerative intervertebral discs,automatic clinical diagnosis and classification,construction of the clinical predictive model and auxiliary operation.(4)In recent years,the research strategy of machine learning has changed to the neural network and deep learning,and the deep learning algorithm with stronger learning ability will be the key to realizing intelligent medical treatment in the future.
ABSTRACT
Objective:To establish a predictive risk model for acute kidney injury (AKI) in acute myocardial infarction (AMI) patients based on machine learning algorithm and compare with a traditional logistic regression model.Methods:It was a retrospective study. The demographic data, laboratory examination, treatment regimen and medication of AMI patients from July 2011 to December 2016 in Beijing Anzhen Hospital, Capital Medical University were collected. The diagnostic criteria of AKI were based on the AKI diagnosis and treatment guidelines published by Kidney Diseases: Improving Global Outcomes in 2012. The selected AMI patients were randomly divided into training set (70%) and internal test set (30%) by simple random sampling. SelectFromModel and Lasso regression models were used to extract clinical parameters as predictors of AKI in AMI patients. Logistic regression model (model A) and machine learning algorithm (model B) were used to establish the risk prediction model of AKI in AMI patients. DeLong method was used to compare the area under the receiver-operating characteristic (ROC) curve ( AUC) between model A and model B for selecting the best model. Results:A total of 6 014 AMI patients were included in the study, with age of (58.4±11.7) years old and 3 414 males (80.5%). There were 674 patients (11.2%) with AKI. There were 4 252 patients (70.7%) in the training set and 1 762 patients (29.3%) in the test set. The selected twelve clinical parameters by the SelectFromModel and Lasso regression models included the number of myocardial infarctions, ST-segment elevation myocardial infarction, ventricular tachycardia, third degree atrioventricular block, decompensated heart failure at admission, admission serum creatinine, admission blood urea nitrogen, admission peak creatine kinase isoenzyme, diuretics, maximum daily dose of diuretics, days of diuretic use and statins. Logistic regression prediction model showed that AUC for the test set was 0.80 (95% CI 0.76-0.84). The machine learning algorithm model obtained AUC in the test set with 0.82 (95% CI 0.78-0.85).There was no significant difference in AUC between the two models ( Z=0.858, P=0.363), and AUC of the machine learning algorithm predictive model was slightly higher than that of the traditional logistic regression model. Conclusions:The prediction effect of AKI risk in AMI patients based on machine learning algorithm is similar to that of traditional logistic regression model, and the prediction accuracy of machine learning algorithm is better. The introduction of machine learning algorithm model may improve the ability to predict AKI risk.
ABSTRACT
AIM: To analyze the influencing factors of postoperative dry eye complication after corneal transplantation and to build a nomogram prediction model.METHODS: Clinical data were collected on 117 patients who underwent corneal transplantation at our hospital from March 2021 to February 2023. They were divided into dry eye group(n=96)and non-dry eye group(n=21)according to whether there was a postoperative dry eye. The risk factors of postoperative complication of dry eye after corneal transplantation were analyzed, the nomogram prediction model for predicting postoperative complication of dry eye after corneal transplantation was constructed, and the internal validation of the model and the prediction efficacy were assessed by calibration curves and decision curves, respectively.RESULTS: Comorbid diabetes mellitus, comorbid sleep disorders, comorbid meibomian gland dysfunction, chronic eye drop abuse, chronic corneal contact lens wear, interleukin-6(IL-6), and tumor necrosis factor-α(TNF-α)were the risk factors for the complication of dry eye after corneal transplantation(P<0.05). The nomogram model predicted a C-index of 0.890(95% CI 0.877-0.903). The nomogram model had a threshold >0.07, and the nomogram model provided higher net clinical benefit than the single index in all cases.CONCLUSION: The nomogram model built in this study based on the factors affecting the complication of dry eye after corneal transplantation has a good predictive value for the complication of dry eye after corneal transplantation.
ABSTRACT
【Objective】 To investigate the factors influencing the length of hospital stays of the acute fatty liver of pregnancy (AFLP) patients, so as to establish the prediction model. 【Methods】 A total of 49 patients diagnosed as AFLP)in ShenZhen People’s Hospital between January 2008 and January 2023 were retrospectively analyzed. According to the median length of hospital stays, the patients were divided into two groups: Group A(n=21)and Group B(n=28). Preoperative general laboratory data, clinical features and postpartum adverse outcomes in both groups were analyzed. Multivariate binary logistic regression was used to analyze the independent factors affecting the length of hospital stays for AFLP, and a prediction model for hospitalization time was established. 【Results】 Comparison between Group B and Group A were as follows: hospital stays(d)(15.5 vs 8), preoperative icterus(%)[16(57.1%)vs 3(14.3%)], thrombin time(TT)(s)(24.2 vs 21.3), prothrombin time(PT)(s)(16.8 vs 15.3), activated partial thromboplastin time(APTT)(s)(52.3 vs 40.7), total bilirubin(TBIL)(μmol/L)(77.2 vs 45.2), indirect bilirubin(IBIL)(μmol/L)(21.2 vs 10), creatinine(Cre)(μmol/L)[(171.97±53.34) vs (131.81±45.06]), TT extension(%)[24(85.7%)vs 11(52.4%)], APTT extension(%)[27(96.4%)vs 7(33.3%)], IBIL elevation(%)[19(67.9%)vs 4(19%)], Cre concentration rise(%)[21(75%)vs 8(38%)], number of postpartum plasma exchange sessions(%)[23(82.1%)vs 5(23.8%)], postpregnancy co-infection phenomenon(%)[21(75%)vs 4(19%)], with Group B significantly higher than Group A. The preoperative platelet count(×109/L)(128 vs 221)and the concentration of fibrinogen(g/L)[0.9 vs 1.6] in Group B were significantly lower than those in Group A. Univariate logistic regression analysis showed that preoperative icterus, postpregnancy co-infection phenomenon, number of postpartum plasma exchange sessions, preoperative TT extension, preoperative APTT extension, Cre concentration rise were influencing factors for the hospital stays of AFLP patients. According to the minimum result of Akaike information criterion, the multivariate binary logistic regression analysis (step-wise selection) showed that the number of postpartum plasma exchange sessions, icterus, preoperative APTT extension were the independent risk factor influencing the hospital stays of AFLP patients, and the logistic regression prediction model was established by incorporating the above three factors. Regularization techniques were further employed in linear regression to address and assess overfitting issues. Additionally, the confidence interval for the estimated effect sizes in each model have been acquired by bootstrapping techniques. 【Conclusion】 Preoperative icterus, preoperative APTT extension(APTT>43s)and the number of postpartum plasma exchange sessions were the independent risk factor influencing the hospital stays of AFLP patients and the logistic regression prediction model with high predictive effectiveness was established successfully.
ABSTRACT
BackgroundThe occurrence rate of dangerous behaviors in patients with severe mental disorders is higher than that of the general population. In China, there is limited research on the prediction of dangerous behaviors in community-dwelling patients with severe mental disorders, particularly in terms of predicting models using data mining techniques other than traditional methods. ObjectiveTo explore the influencing factors of dangerous behaviors in community-dwelling patients with severe mental disorders and testing whether the classification decision tree model is superior to the Logistic regression model. MethodsA total of 11 484 community-dwelling patients with severe mental disorders who had complete follow-up records from 2013 to 2022 were selected on December 2023. The data were divided into a training set (n=9 186) and a testing set (n=2 298) in an 8∶2 ratio. Logistic regression and classification decision trees were separately used to establish predictive models in the training set. Model discrimination and calibration were evaluated in the testing set. ResultsDuring the follow-up period, 1 115 cases (9.71%) exhibited dangerous behaviors. Logistic regression results showed that urban residence, poverty, guardianship, intellectual disability, history of dangerous behaviors, impaired insight and positive symptoms were risk factors for dangerous behaviors (OR=1.778, 1.459, 2.719, 1.483, 3.890, 1.423, 2.528, 2.124, P<0.01). Being aged ≥60 years, educated, not requiring prescribed medication and having normal social functioning were protective factors for dangerous behaviors (OR=0.594, 0.824, 0.422, 0.719, P<0.05 or 0.01). The predictive effect in the testing set showed an area under curve (AUC) of 0.729 (95% CI: 0.692~0.766), accuracy of 70.97%, sensitivity of 59.71%, and specificity of 72.05%. The classification decision tree results showed that past dangerous situations, positive symptoms, overall social functioning score, economic status, insight, household registration, disability status and age were the influencing factors for dangerous behaviors. The predictive effect in the testing set showed an AUC of 0.721 (95% CI: 0.705~0.737), accuracy of 68.28%, sensitivity of 64.46%, and specificity of 68.60%. ConclusionThe classification decision tree does not have a greater advantage over the logistic regression model in predicting the risk of dangerous behaviors in patients with severe mental disorders in the community. [Funded by Chengdu Medical Research Project (number, 2020052)]
ABSTRACT
In medical research,predictive models have been widely used to predict disease progression and identify high-risk populations in advance, especially in the prevention and diagnosis of chronic diseases. In ophthalmology, the predictive and diagnostic models for fundus diseases such as age-related macular degeneration and diabetic retinopathy have demonstrated expert-level accuracy. However, the application of predictive models is still in the exploratory stage as for myopia prevention and control. The establishment of a predictive model is helpful to identify the high-risk myopic children in advance, so that preventive measures such as adequate outdoor activities and reducing near work can be taken in time, which is of great significance to prevent or slow down the occurrence and development of myopia. Because the mechanism of myopia has not been fully elucidated, there are still challenges and limitations in the selection of application objects, predictors and predictive outcomes. This paper reviews the research progress of different types of myopia predictive models in order to provide reference for further development and improvement.
ABSTRACT
Objectives:To explore the risk factors related to the prolonged postoperative length of hospital stay(LOS)in patients after spinal tuberculosis lesion removal and fusion with internal fixation,and to construct a nomogram prediction model,so as to provide a theoretical basis for the enhanced recovery management of spinal tuberculosis patients.Methods:The clinical data of 142 patients with spinal tuberculosis who underwent lesion removal and fusion with internal fixation in the Department of Orthopedics of the Affiliated Hospital of Zunyi Medical University between December 2018 and June 2023 were retrospectively analyzed.The patients were randomly divided into modeling group(n=96)and validation group(n=46)in a 2∶1 ratio.Setting the postoperative LOS>21d as the outcome variable for prolonged LOS,and taking age,gender,alcohol history,smoking history,hypertension,coronary heart disease,diabetes,anemia,postoperative hypoproteinemia,spinal cord injury,tuberculosis in other parts,bone destruction,blood transfusion,removal time of drainage,postoperative complications,operative time,blood loss,preoperative American Society of Anesthesiologists(ASA)score,postoperative ASA score,surgical incision length,pus formation,chemotherapy before surgery,and chemotherapy regimens as independent variables to develop univariate logistic regression model.The risk factors screened after univariate analysis were included for multivariate logistic regression model to determine the independent risk factors for LOS>21d after lesion removal and fusion with internal fixation in patients with spinal tuberculosis and to construct a predictive model for risk factors.The area under the curve(AUC)of receiver operating characteristics(ROC)curve was used to assess the the differentiation of the model;Calibration curve was used to assess the calibration situation of the model;Decision curve analysis(DCA)was used to assess the clinical value and influence of the model on actual decision-making process.Data of validation group was applied to draw ROC curve and calibration curve for external verification.Results:Univariate and multivariate analyses revealed that age(OR=1.040,95%CI:1.011-1.069),tuberculosis at other sites(OR=2.867,95%CI:1.157-7.106),and preoperative ASA score(OR=1.543,95%CI:1.015-2.347)were the independent risk factors for prolonged postoperative hospitalization in patients with spinal tuberculosis after lesion removal and fusion with internal fixation.The AUC of ROC curves of modeling group and validation group were 0.767(95%CI:0.671-0.863)and 0.720(95%CI:0.569-0.871),respectively,suggesting the predictive model had good predictive efficiency.The results of the calibration curve analysis demonstrated that the actual curve roughly resembled the ideal curve,and DCA curve revealed that the nomogram had superior clinical benefits.Conclusions:The spinal tuberculosis patients who are at older age,combined with other sites of tuberculosis,and with high preoperative ASA score are prone to prolonged LOS after lesion removal and fusion with internal fixation,and the risk prediction nomogram model developed accordingly has great predictive efficiency.
ABSTRACT
Objectives:To analyze the risk factors related to infection after posterior lumbar interbody fusion(PLIF)by random forest algorithm and develop a prediction model,providing a certain reference for clinical prevention of surgical site infection(SSI)after PLIF.Methods:A retrospective study was conducted on the masked data of patients hospitalized for PLIF in the spinal surgery department of some third-level grade A hospitals in Beijing municipality and Hebei Province from June 2019 to June 2021 provided by Beijing Zhongwei Cloud Medical Data Analysis and Application Technology Research Institute through data processing and analysis.The classification data were analyzed and compared between SSI group and non-SSI group to obtain variables that significantly impacted the postoperative infection.SPSS Modeler 20 system was used as the tool for model development,and random forest algorithm was applied to analyze,obtaining the patient characteristics of postoperative infection,namely the infection model.Results:A total of 8,764 patients were included in study,and 373 patients were diagnosed with SSI,with an incidence rate of 4.4%(95%CI:2.2%to 6.5%).After statistical analysis,six variables,including obesity,ASA Ⅲ and above,prolonged operative time,chronic heart disease,diabetes and renal dysfunction,were independently associated with SSI.Classification with a random forest model yielded a high accuracy of 90.6%.The characteristics of patients prone to infection after PLIF(two models of infection)was:[(BMI=1)and(SD=1)and(ASA=1)and(RI=1)]or[(BMI=0)and(SD=1)and(DM=1)and(RI=1)].Conclusions:The random forest algorithm applied in this study could obtain an average accuracy of 90.6%,and two infection models were obtained as:(1)Patients with obesity,renal insufficiency,ASA grade Ⅲ or above,and operative time≥3h;(2)Patients who are not obese,but with diabetes,renal insufficiency,and the operative time ≥3h.
ABSTRACT
Objective:To investigate the predictive value of the model based on soluble T cell immunogloblulin and mucin domain-containing protein 3 (sTIM3) for the progression of severe acute pancreatitis (SAP) in patients with acute pancreatitis (AP).Methods:A retrospective cohort study was conducted. The AP patients admitted to Changzhou First People's Hospital and Changzhou Second People's Hospital from June 1, 2020 to June 30, 2022 were enrolled. Mild AP (MAP) and moderately severe AP (MSAP) patients were classified as non-SAP group, and SAP patients were classified as SAP group according to the progression of AP patients during hospitalization. The basic data, blood biological indicators, serum sTIM3 level, bedside index for severity in acute pancreatitis (BISAP), acute physiology and chronic health evaluation Ⅱ (APACHE Ⅱ) score, modified computed tomography severity index (MCTSI) score within 48 hours of admission, and prognosis indicators were collected. Multivariate Logistic regression analysis was conducted to analyze the risk factors of the progression of SAP in patients with AP during hospitalization. Based on the results of multivariate analysis and the best parameters selected based on the minimal Akaike information criterion (AIC), the SAP prediction model based on sTIM3 was constructed. The receive operator characteristic curve (ROC curve) was plotted to analyze the predictive efficacy of the model.Results:A total of 99 AP patients were enrolled, 80 patients in the non-SAP group and 19 patients in the SAP group. Compared with the non-SAP group, body mass index (BMI), drinking history ratio, heart rate (HR), respiration rate (RR), white blood cell count (WBC), red blood cell count (RBC), C-reactive protein (CRP), alanine aminotransferase (ALT), serum creatinine (SCr), procalcitonin (PCT), interleukin-6 (IL-6), sTIM3, BISAP score, APACHEⅡ score and MCTSI score were significantly increased, and pulse oxygen saturation (SpO 2), direct bilirubin (DBil) and IL-10 were significantly decreased. The length of intensive care unit (ICU) stay and total length of hospital stay of patients in the SAP group were significantly longer than those in the non-SAP group [length of ICU stay (days): 1.0 (0, 1.5) vs. 0 (0, 0), total length of hospital stay (days): 17.11±9.39 vs. 8.40±3.08, both P < 0.01]. Multivariate Logistic regression analysis showed that HR [odds ratio ( OR) = 1.059, 95% confidence interval (95% CI) was 1.010-1.110, P = 0.017], DBil ( OR = 0.981, 95% CI was 0.950-0.997, P = 0.043), and sTIM3 ( OR = 1.002, 95% CI was 1.001-1.003, P = 0.027) were independent risk factors for predicting the progression of SAP in patients with AP, and the SAP prediction model based on sTIM3 was constructed: Logit( P) = -14.602+0.187×BMI+0.057×HR+0.006×CRP-0.020×DBil+0.002×sTIM3. ROC curve analysis showed that among the aforementioned single factor quantitative indicators, IL-6 was the most effective in predicting the progression of AP patients to SAP during hospitalization, but the predictive performance of prediction model based on the sTIM3 was significantly better than IL-6 [area under the ROC curve (AUC) and 95% CI: 0.957 (0.913-1.000) vs. 0.902 (0.845-0.958), P < 0.05]. Conclusion:The model based on serum sTIM3 demonstrated good predictive value for the progression of SAP in patients with AP.
ABSTRACT
Objective Differential genes related to prognosis of endometrial carcinoma(EC)were screened and prognostic models were constructed.Methods Gene Expression data of EC and normal control samples were downloaded from The Cancer Genome Atlas(TCGA)database and Gene Expression Omnibus(GEO)dataset GSE63678 to screen out common differential genes.LASSO regression analysis was used to screen out the genes with prognostic significance and construct prognostic characteristics.EC patients with complete information were obtained from the TCGA database and randomly divided into the training group and the validation group in a ratio of 1:1.In the training group,survival curves were constructed based on prognostic characteristics.Functional annotation and pathway enrichment analysis were conducted using gene ontology(GO)analysis and Kyoto Encyclopedia of Genes and Genomes(KEGG)analysis.Combined with prognostic features and clinical risk factors,a calibration curve and C-index were used to evaluate the performance of the histogram.Finally,use a verification group for validation.Results A total of 4800 and 257 differentially expressed genes were screened from TCGA and GEO databases respectively,of which 73 up-regulated genes and 52 down-regulated genes were co-expressed.6 prognostic genes(ORMDL2,BNC2,TTK,MAMLD1,KCTD7 and DCLK2)were screened out by LASSO regression analysis.The survival curve showed that the overall survival of patients in the high-risk group was significantly lower than that in the low-risk group(P<0.01).GO analysis and KEGG results exhibited that prognostic signature was associated with cell cycle.The nomogram showed powerful predictive ability in the training and validation groups.Conclusion We constructed a predictive model based on prognostic genes,which can accurately predict the prognosis of patients with EC and provide new theoretical support for clinical diagnosis and treatment.
ABSTRACT
Introducción: El diagnóstico precoz de la crisis vasoclusiva (CVO) que afecta a pacientes con drepanocitosis resulta un tema no resuelto en la actualidad. No se ha encontrado en la literatura evidencia de modelos que puedan establecer tempranamente índices de riesgo de la CVO para la toma de una conducta terapéutica oportuna en estos pacientes. Objetivo: Establecer índices de riesgo en pacientes con drepanocitosis, a partir de la formulación de un modelo predictivo del estado vasoclusivo. Métodos: A partir de un estudio analí tico transversal de casos y controles, realizado en el Centro Hematológico de Santiago de Cuba, se formuló a través de un análisis discriminante, un modelo predictivo del estado de CVO. Se usaron estadígrafos de dispersión (media y desviación estándar) para el establecimiento de índices de riesgo sustentados en él. Resultados: Se formuló un modelo predictivo del estado de CVO que incluyó biomarcadores del estado redox como predictores significativos en el paciente con drepanocitosis. El modelo sustentó los índices de riesgo, estratificados en 3 categorías (riesgo menor, moderado y mayor) que fueron asignados a los pacientes y posibilitó su adecuada clasificación. Conclusiones: El diseño de un modelo predictivo de CVO y el establecimiento de índices de riesgo en pacientes con drepanocitosis permitió una mejor evaluación. La nueva herramienta diagnóstica que se propone resultaría de gran utilidad en los servicios de Hematología, al facilitar una mejor valoración del estado del paciente con drepanocitosis y un tratamiento profiláctico oportuno que minimice las complicaciones asociadas a este estado.
Introduction: The early diagnosis of vasooclusive crisis affecting sickle cell patients is currently an unresolved issue. In the reviewed literature no models have been found able to establish early risk indices of vasooclusive crisis for taking a timely therapeutic behavior in these patients. Objective: To establish risk indices in sickle cell patients based on the formulation of a predictive model of vasoocclusive state. Methods: Based on a cross-sectional case-control analytic study conducted at the Hematological Center of Santiago de Cuba, a predictive model of VOC status was formulated through a discriminant analysis. Dispersion statistics (mean and standard deviation) were used to establish risk indices based on it. Results: A predictive model of the state of VOC that included biomarkers of the redox state as significant predictors of it in sickle cell patients was formulated. The model supported the risk indices, stratified into 3 categories (lower, moderate and higher risk) that were assigned to the patients and allowed an adequate classification of them. Conclusions: The design of a predictive model of VOC and the establishment of risk indices in sickle cell patients allowed a better evaluation of them. The new diagnostic tool proposed in the study would be very useful in the Hematology services, by facilitating a better assessment of the sickle cell patient's condition and a timely prophylactic treatment that minimizes the complications associated with this state.
Subject(s)
Humans , ForecastingABSTRACT
Introducción: La fibrilación auricular es la arritmia recurrente más habitual en la práctica clínica. Su prevalencia se multiplica en la población actual y tiene diferentes causas fisiopatológicas que la convierten en una pandemia mundial. Objetivos: Diseñar un modelo predictivo de fracaso de la terapia eléctrica en pacientes con fibrilación auricular paroxística. Métodos: Se realizó un estudio de casos y controles, con 33 casos y 66 controles. Variables predictoras: edad, fracción de eyección ≤ 40 por ciento, volumen de aurícula izquierda ≥ 34 mL/m2. A partir de la regresión logística se obtuvo un modelo en el que fueron incluidos el valor predictivo positivo, valor predictivo negativo, la sensibilidad y especificidad. Resultados: Los factores de riesgo predictores fueron: edad ≥ 55 años (p= 0,013; odds ratio (OR)= 3,58; intervalo de confianza -IC- 95 por ciento: 1,33-9,67); la fracción de eyección del ventrículo izquierdo (FEVI) ≤ 40 por ciento se observó en 20 pacientes (22,7 por ciento) (p= 0,004; OR= 4,45; IC95 por ciento: 1,54-12,8); presión de aurícula izquierda elevada, volumen de aurícula izquierda elevado (p= 0,004; OR= 3,11; IC95 por ciento: 1,24-8,77), según el modelo de regresión logística. Se realizó la validación interna por división de datos; se confirmó que el modelo pronostica bien los que van a tener éxito en el resultado terapéutico. Conclusiones: El modelo predictivo elaborado está compuesto por los predictores edad > 55 años, FEVI; volumen de aurícula izquierda; presenta un buen ajuste y poder discriminante, sobre todo valor predictivo positivo(AU)
Introduction: Atrial fibrillation is the most common recurrent arrhythmia in clinical practice. Its prevalence is multiplying in the current population and has different pathophysiological causes that make it a global pandemic. Objectives: To design a predictive model for failure of electrical therapy in patients with paroxysmal atrial fibrillation. Methods: A case-control study was carried out with 33 cases, and 66 controls. Predictor variables: age, ejection fraction ≤ 40 percent, left atrial volume ≥ 34 mL/m2. From logistic regression, a model was obtained in which the positive predictive value, negative predictive value, sensitivity and specificity were included. Results: The predictive risk factors were: age ≥ 55 years (p= 0.013; odds ratio (OR)= 3.58; 95 percent confidence interval -CI-: 1.33-9.67); left ventricular ejection fraction (LVEF) ≤ 40 percent was observed in 20 patients (22.7 percent) (p= 0.004; OR= 4.45; 95 percent CI: 1.54-12.8); elevated left atrial pressure, elevated left atrial volume (p= 0.004; OR= 3.11; 95 percent CI: 1.24-8.77), according to the logistic regression model. Internal validation was carried out by data division; It was confirmed that the model predicts very well those who will be successful in the therapeutic result. Conclusions: The predictive model developed is composed of the predictors age > 55 years, LVEF; left atrial volume; It presents a good fit and discriminating power, especially positive predictive value(AU)