OBJECTIVE: To explore the risk factors affecting the prognosis of patients with sepsis complicated with acute pulmonary embolism, and to construct and validate a nomogram predictive model for in-hospital mortality risk. METHODS: Based on the American Medical Information Mart for Intensive Care (MIMIC-III, MIMIC-IV) databases, the data were collected on patients with sepsis complicated with acute pulmonary embolism from 2001 to 2019, including baseline characteristics, and vital signs, disease scores, laboratory tests within 24 hours of admission to the intensive care unit (ICU), and interventions. In-hospital mortality was the outcome event. The total samples were divided into training and testing sets in a 7:3 ratio by random sampling. Univariate Cox regression analysis was used to verify the impact of all variables on the risk of in-hospital mortality, thereby screen potential influencing factors. Subsequently, a stepwise bi-directional regression method was applied to select factors one by one, leading to the construction of a nomogram prediction model. Collinearity testing was used to demonstrate the absence of strong multicollinearity among the influencing factors in the nomogram prediction model. The discrimination of the nomogram model, sequential organ failure assessment (SOFA), and simplified pulmonary embolism severity index (sPESI) was evaluated using C-index in the test set. Receiver operator characteristic curve (ROC curve) was drawn to evaluate the predictive value of various models for in-hospital mortality in patients with sepsis complicated with acute pulmonary embolism. RESULTS: A total of 562 patients with sepsis complicated with acute pulmonary embolism were included, including 393 in the training set and 169 in the testing set. Univariate Cox regression analysis showed that 30 factors associated with in-hospital mortality in patients with sepsis complicated with acute pulmonary embolism. Through stepwise bi-directional regression, 12 variables were ultimately selected, including gender, presence of malignant tumors, body temperature, red cell distribution width (RDW), blood urea nitrogen (BUN), serum potassium, prothrombin time (PT), 24-hour urine output, mechanical ventilation, vasoactive drugs, warfarin use, and sepsis-induced coagulopathy (SIC). Collinearity testing indicated no strong multicollinearity among the influencing factors [all variance inflation factor (VIF) > 10]. A nomogram model was constructed using the 12 variables mentioned above. The nomogram model predicted the C-index and its 95% confidence interval (95%CI) of in-hospital mortality in patients with sepsis complicated with acute pulmonary embolism better than SOFA score and sPESI [0.771 (0.725-0.816) vs. 0.579 (0.519-0.639), 0.608 (0.554-0.663)]. The ROC curve showed that the area under the curve (AUC) and its 95%CI of the nomogram model were higher than those of the SOFA score and sPESI [0.811 (0.766-0.857) vs. 0.630 (0.568-0.691), 0.623 (0.566-0.680)]. These findings were consistently replicated in the internal validation of the testing set. In both the training and testing sets, Delong's test showed that the AUC of the nomogram model was significantly higher than the SOFA score and sPESI (both P < 0.05). CONCLUSION The nomogram model demonstrated good predictive effectiveness for the risk of in-hospital mortality in patients with sepsis complicated with acute pulmonary embolism, enabling clinicians to predict mortality risk in advance and take timely interventions to reduce mortality.
Humans ; Pulmonary Embolism/mortality* ; Hospital Mortality ; Nomograms ; Sepsis/complications* ; Prognosis ; Risk Factors ; Intensive Care Units ; Male ; Female ; Middle Aged ; Aged

OBJECTIVE: To construct machine learning prediction model for sepsis-associated encephalopathy (SAE), and analyze the application value of the model on early identification of SAE risk in elderly septic patients. METHODS: Patients aged over 60 years with a primary diagnosis of sepsis admitted to intensive care unit (ICU) from 2008 to 2023 were selected from Medical Information Mart for Intensive Care-IV 2.2 (MIMIC-IV 2.2). Demographic variables, disease severity scores, comorbidities, interventions, laboratory indicators, and hospitalization details were collected. Key factors associated with SAE were identified using univariate Logistic regression analysis. The data were randomly divided into training and validation sets in a 7 : 3 ratio. Multivariable Logistic regression analysis was conducted in the training set and visualized using a nomogram model for prediction of SAE. The discrimination of the model was evaluated in the validation set using the receiver operator characteristic curve (ROC curve), and its calibration was assessed using calibration curve. Furthermore, multiple machine learning algorithms, including multi-layer perceptron (MLP), support vector machine (SVM), naive bayes (NB), gradient boosting machine (GBM), random forest (RF), and extreme gradient boosting (XGB), were constructed in the training set. Their predictive performance was subsequently evaluated on the validation set. Taking the XGB model as an example, the interpretability of the model through the SHapley Additive exPlanations (SHAP) algorithm was enhanced to identify the key predictive factors and their contributions. RESULTS: A total of 2 204 septic patients were finally enrolled, of whom 840 developed SAE (38.1%). A total of 21 variables associated with SAE were screened through univariate Logistic regression analysis. Multivariable Logistic regression analysis showed that endotracheal intubation [odds ratio (OR) = 0.40, 95% confidence interval (95%CI) was 0.19-0.88, P < 0.001], oxygen therapy (OR = 0.76, 95%CI was 0.53-0.95, P = 0.023), tracheotomy (OR = 0.20, 95%CI was 0.07-0.53, P < 0.001), continuous renal replacement therapy (CRRT; OR = 0.32, 95%CI was 0.15-0.70, P < 0.001), cerebrovascular disease (OR = 0.31, 95%CI was 0.16-0.60, P < 0.001), rheumatic disease (OR = 0.44, 95%CI was 0.19-0.99, P < 0.001), male (OR = 0.68, 95%CI was 0.54-0.86, P = 0.001), and maximum anion gap (AG; OR = 0.95, 95%CI was 0.93-0.97, P < 0.001) were associated with an decreased probability of SAE, and age (OR = 1.05, 95%CI was 1.03-1.06, P < 0.001), acute physiology score III (APSIII; OR = 1.02, 95%CI was 1.01-1.02, P < 0.001), Oxford acute severity of illness score (OASIS; OR = 1.04, 95%CI was 1.03-1.06, P < 0.001), and length of hospital stay (OR = 1.01, 95%CI was 1.01-1.02, P < 0.001) were associated with an increased probability of SAE. A nomogram model was constructed based on these variables. In the validation set, ROC curve analysis showed that the model achieved an area under the ROC curve (AUC) of 0.723, and the calibration curve showed good consistency between the predicted probability of the model and the observed probability. Among the machine learning algorithms, including MLP, SVM, NB, GBM, RF, and XGB, the SVM model and RF model demonstrated relatively good predictive performance, with AUC of 0.748 and 0.739, respectively, and the sensitivity was both exceeding 85%. The predictive performance of the XGB model was explained through SHAP analysis, and the results indicated that APSIII score (SHAP value was 0.871), age (SHAP value was 0.521), and OASIS score (SHAP value was 0.443) were important factors affecting the predictive performance of the model. CONCLUSIONS The machine learning-based SAE prediction model exhibits good predictive capability and holds significant application value for the early identification of SAE risk in elderly septic patients.
Humans ; Machine Learning ; Aged ; Sepsis-Associated Encephalopathy ; Sepsis/complications* ; Intensive Care Units ; Logistic Models ; Middle Aged ; Male ; ROC Curve ; Female ; Bayes Theorem ; Nomograms ; Support Vector Machine ; Algorithms

Objective:To explore the association between albumin-corrected serum calcium (ACSC) levels and 30-day all-cause mortality in patients hospitalized for community-acquired bacterial pneumonia (CABP).Methods:A secondary analysis was conducted on 1 899 patients with CABP from a Norwegian cohort study. The relationship between baseline ACSC levels and 30-day mortality was assessed using multivariable logistic regression models, adjusted for potential confounders.Results:A significant positive correlation was found between ACSC levels and 30-day mortality risk after adjusting for confounding variables ( OR=1.95, 95% CI=1.48-2.58). When ACSC levels were categorized into tertiles (T1-T3), a trend analysis revealed that the T2 and T3 groups had significantly higher mortality risks compared to the lowest tertile (T1), with odds ratios ( OR) and 95% confidence intervals ( CI) of 1.52(95% CI: 0.97-2.38) and 2.21 (95% CI: 1.44-3.39), respectively ( P for trend <0.001). Subgroup analyses demonstrated no significant interactions across predefined subgroups (all P for interaction > 0.05). Conclusions:In patients with CABP and admission ACSC levels of ≥8.6 mg/dL, higher ACSC levels were positively associated with an increased risk of 30-day mortality. These findings highlight the potential prognostic value of ACSC levels in CABP patients.

Objective To explore the failure criterion that can accurately simulate the tensile and compressive fracture of hollow cortical bone structure.Methods Based on the previous compression and bending experimental data,the predicted results using different failure criteria were compared to determine the simulation accuracy.Results Under the compressive load,the differences in the fracture load between the simulations using the equivalent and invariant strain failure criteria and the experiment were less than 5％,indicating that these two failure criteria were suitable for predicting the cortical bone failure;under the bending load,the differences in the fracture load between the simulations using the equivalent and invariant strain failure criteria and the experiment were less than 5％,indicating that these two failure criteria could accurately predict the failure process.Conclusions The prediction accuracy using different failure criteria mainly depends on whether the strain growth rate conforms to the actual bone deformation.Too fast or too low strain growth rate will lead to premature or delayed structural fracture.The fracture simulation adopted in this study is suitable for most cortical bone structures,and can be used to determine the suitable failure criterion under different loads,so as to assist in obtaining the strength limit of cortical bone in various parts and provide data support for improving the simulation accuracy and grasping the condition of fracture occurrence in clinical practice.

Objective To explore the failure criterion that can accurately simulate the tensile and compressive fracture of hollow cortical bone structure.Methods Based on the previous compression and bending experimental data,the predicted results using different failure criteria were compared to determine the simulation accuracy.Results Under the compressive load,the differences in the fracture load between the simulations using the equivalent and invariant strain failure criteria and the experiment were less than 5％,indicating that these two failure criteria were suitable for predicting the cortical bone failure;under the bending load,the differences in the fracture load between the simulations using the equivalent and invariant strain failure criteria and the experiment were less than 5％,indicating that these two failure criteria could accurately predict the failure process.Conclusions The prediction accuracy using different failure criteria mainly depends on whether the strain growth rate conforms to the actual bone deformation.Too fast or too low strain growth rate will lead to premature or delayed structural fracture.The fracture simulation adopted in this study is suitable for most cortical bone structures,and can be used to determine the suitable failure criterion under different loads,so as to assist in obtaining the strength limit of cortical bone in various parts and provide data support for improving the simulation accuracy and grasping the condition of fracture occurrence in clinical practice.

BACKGROUND:Critical energy release rate is a global fracture parameter that could be measured during the failing process,and its value may change under different failure modes even in the same structure. OBJECTIVE:To propose an approach to predict the critical energy release rate in the femoral cortical bone structure under different failure modes. METHODS:Three-point bending and axial compression experiments and the corresponding fracture simulations were performed on the rat femoral cortical bone structures.Different critical energy release rates were repeatedly assigned to the models to perform fracture simulation,and the predicted load-displacement curves in each simulation were compared with the experimental data to back-calculate the critical energy release rate.The successful fit was that the differences in the fracture parameters between the predicted and experimental results were less than 5%. RESULTS AND CONCLUSION:(1)The results showed that the cortical bone structure occurred tensile open failure under three-point bending load,and the predicted critical energy release rate was 0.16 N/mm.(2)The same cortical bone structure occurred shear open failure under axial compression load,and the predicted critical energy release rate was 0.12 N/mm,which indicates that the critical energy release rate of the same cortical bone structure under different failure modes was different.(3)A comprehensive analysis from the perspectives of material mechanical properties and damage mechanism was conducted to reveal the reasons for the differences in the critical energy release rate in the cortical bone structure under different failure modes,which provided a theoretical basis for the measurement of the energy release rate and the accurate fracture simulation.

Objective To predict the tissue-level failure strain of the cortical bone and discuss the effects of different running speeds on the mechanical properties of rat femoral cortical bone.Methods The threshold for cortical bone tissue-level failure strain was assigned,and fracture simulation under three-point bending was performed on a rat femoral finite element model.The predicted load-displacement curves in each simulation were compared and fitted with the experimental data to back-calculate the tissue-level failure strain.Results The cortical bone tissue-level failure strains at different running speeds were statistically different,which indicated that different running speeds had certain impacts on the micromechanical properties of the cortical bone structures.At a running speed of 12 m/min,the cortical bone structure expressed the greatest tissue-level failure strain,and at a running speed of 20 m/min,the cortical bone structure expressed the lowest tissue-level failure strain.Conclusions Based on the changing trends of tissue-level failure strain and in combination with the changes in macro-level failure load and tissue-level elastic modulus of cortical bone structures,the effects of different running speeds on the mechanical properties of cortical bone structures were discussed in this study.The appropriate running speed for improving the mechanical properties of the cortical bone was explored,thereby providing a theoretical basis for improving bone strength through running exercises.

Objective To predict the micro-level energy release rate in the rat femoral cortical bone and investigate the variation in the micro-level energy release rate with age.Methods Based on previous experimental data and numerical simulation of fracture modes for cortical bone,load-displacement curves and fracture modes measured by simulation and experiment were compared,and the micro-level energy release rates of rat femoral cortical bone at different months were predicted by back-calculation.Results It was predicted that the micro-level energy release rate of rat femoral cortical bone at 1-,3-,5-,7-,9-,11-,and 15-month age was 0.08-0.12,0.12-0.14,0.15-0.19,0.25-0.28,0.23-0.25,0.19-0.22,and 0.13-0.16 N/mm,respectively.Conclusions The decrease in the microlevel energy release rate with increasing age led to a decreasing failure load,indicating that the microlevel energy release rate is one of the main factors determining fracture occurrence;however,no significant decrease was observed at the time of fracture,indicating that the microlevel energy release rate was not linearly proportional to the fracture time.These results can help explain the mechanism of cortical bone fractures at the clinical level.

Objective:To develop and evaluate a nomogram prediction model for the 3-month mortality risk of patients with sepsis-associated acute kidney injury (S-AKI).Methods:Based on the American Medical Information Mart for Intensive Care-Ⅳ (MIMIC-Ⅳ), clinical data of S-AKI patients from 2008 to 2021 were collected.Initially, 58 relevant predictive factors were included, with all-cause mortality within 3 months as the outcome event. The data were divided into training and testing sets at a 7∶3 ratio. In the training set, univariate Logistic regression analysis was used for preliminary variable screening. Multicollinearity analysis, Lasso regression, and random forest algorithm were employed for variable selection, combined with the clinical application value of variables, to establish a multivariable Logistic regression model, visualized using a nomogram. In the testing set, the predictive value of the model was evaluated through internal validation. The receiver operator characteristic curve (ROC curve) was drawn, and the area under the curve (AUC) was calculated to evaluate the discrimination of nomogram model and Oxford acute severity of illness score (OASIS), sequential organ failure assessment (SOFA), and systemic inflammatory response syndrome score (SIRS). The calibration curve was used to evaluate the calibration, and decision curve analysis (DCA) was performed to assess the net benefit at different probability thresholds.Results:Based on the survival status at 3 months after diagnosis, patients were divided into 7?768 (68.54%) survivors and 3?566 (31.46%) death. In the training set, after multiple screenings, 7 variables were finally included in the nomogram model: Logistic organ dysfunction system (LODS), Charlson comorbidity index, urine output, international normalized ratio (INR), respiratory support mode, blood urea nitrogen, and age. Internal validation in the testing set showed that the AUC of nomogram model was 0.81 [95% confidence interval (95% CI) was 0.80-0.82], higher than the OASIS score's 0.70 (95% CI was 0.69-0.71) and significantly higher than the SOFA score's 0.57 (95% CI was 0.56-0.58) and SIRS score's 0.56 (95% CI was 0.55-0.57), indicating good discrimination. The calibration curve demonstrated that the nomogram model's calibration was better than the OASIS, SOFA, and SIRS scores. The DCA curve suggested that the nomogram model's clinical net benefit was better than the OASIS, SOFA, and SIRS scores at different probability thresholds. Conclusions:A nomogram prediction model for the 3-month mortality risk of S-AKI patients, based on clinical big data from MIMIC-Ⅳ and including seven variables, demonstrates good discriminative ability and calibration, providing an effective new tool for assessing the prognosis of S-AKI patients.

Objective To investigate the value of derived neutrophil to lymphocyte ratio(dNLR)as a pre-dictive biomarker for immunotherapy response in advanced non-small cell lung cancer(NSCLC).Methods A total of 92 patients with advanced NSCLC who received anti-programmed cell death receptor(PD-1)combined therapy in the hospital from August 2018 to December 2019 were selected as the research objects.Peripheral blood samples were collected within 24 h before immunotherapy,complete blood cell count was measured,and dNLR was calculated.Patients with advanced NSCLC were treated with PD-1 inhibitors or combination regi-mens,and the response to immunotherapy was evaluated by objective response rate(ORR)and disease control rate(DCR).The receiver operating characteristic(ROC)curve was used to analyze the predictive value of dN-LR for the diagnosis and response to immunotherapy in advanced NSCLC.Multivariate Logistic regression model was used to analyze the relationship between dNLR and immunotherapy response in advanced NSCLC.Kaplan-Meier survival curve and Log-Rank test were used to analyze the overall survival(OS),progression-free survival(PFS)and disease-specific survival(DSS)of the low dNLR group and the high dNLR group.Re-sults The ORR and DCR of advanced NSCLC patients after immunotherapy were 32.61％and 65.22％,re-spectively,and the PFS and OS were 17.0(8.5,25.5)and 24.0(12.7,36.1)months,respectively.The dNLR of DCR group was lower than that of non-DCR group(P＜0.001).The dNLR of ORR group was lower than that of non-ORR group(P＜0.001).The area under the curve of dNLR for predicting DCR or ORR was 0.897(95％CI 0.829-0.965)and 0.874(95％CI 0.795-0.953),respectively.Multivariate Logistic regres-sion analysis showed that dNLR≥2.28 increased the risk of non-response to immunotherapy,and this inde-pendent relationship still existed after further adjustment for objective confounding factors(P＜0.05).Sur-vival curve results showed that patients with high dNLR had significantly shorter PFS,OS,and DSS(P＜0.05).Multivariate Cox regression analysis showed that high dNLR was an independent factor affecting the poor prognosis of patients with advanced NSCLC(P＜0.05).Conclusion High dNLR advanced NSCLC pa-tients are more difficult to benefit from immune therapy,and prognosis is worse.dNLR is promising as a pre-dictive biomarker for immunotherapy response in advanced NSCLC.