Net Clinical Benefit Research Articles

Development and validation of a nomogram model for prolonged length of stay in spinal fusion patients: a retrospective analysis

ObjectiveTo develop a nomogram model for the prediction of the risk of prolonged length of hospital stay (LOS) in spinal fusion patients.MethodsA retrospective cohort study was carried out on 6272 patients who had undergone spinal fusion surgery. Least absolute shrinkage and selection operator (LASSO) regression was performed on the training sets to screen variables, and the importance of independent variables was ranked via random forest. In addition, various independent variables were used in the construction of models 1 and 2. A receiver operating characteristic curve was used to evaluate the models’ predictive performance. We employed Delong tests to compare the area under the curve (AUC) of the different models. Assessment of the models’ capability to improve classification efficiency was achieved using continuous net reclassification improvement (NRI) and integrated discrimination improvement (IDI). The Hosmer–Lemeshow method and calibration curve was utilised to assess the calibration degree, and decision curve to evaluate its clinical practicality. A bootstrap technique that involved 10 cross-validations and was performed 10,000 times was used to conduct internal and external validation. The were outcomes of the model exhibited in a nomogram graphics. The developed nomogram was validated both internally and externally.ResultsModel 1 was identified as the optimal model. The risk factors for prolonged LOS comprised blood transfusion, operation type, use of tranexamic acid (TXA), diabetes, electrolyte disturbance, body mass index (BMI), surgical procedure performed, the number of preoperative diagnoses and operative time. The diagnostic performance of the nomogram model was satisfactory, with AUC values of 0.784 and 0.795 for the internal and external validation sets, respectively. Model discrimination was favourable in both the internal (C-statistic, 0.811) and external (C-statistic, 0.814) validation sets. Calibration curve and Hosmer-Lemeshow test showed acceptable agreement between predicted and actual results. The decision curve shows that the model provides net clinical benefit within a certain decision threshold range.ConclusionsThis study developed and validated a nomogram to identify the risk of prolonged LOS in spinal fusion patients, which may help clinicians to identify high-risk groups at an early stage. Predictors identified included blood transfusion, operation type, use of TXA, diabetes, electrolyte disturbance, BMI, surgical procedure performed, number of preoperative diagnoses and operative time.

Development of a neoadjuvant chemotherapy efficacy prediction model for nasopharyngeal carcinoma integrating magnetic resonance radiomics and pathomics: a multi-center retrospective study

ObjectiveThis study aimed to develop and validate a predictive model for assessing the efficacy of neoadjuvant chemotherapy (NACT) in nasopharyngeal carcinoma (NPC) by integrating radiomics and pathomics features using a particle swarm optimization-supported support vector machine (PSO-SVM).MethodsA retrospective multi-center study was conducted, which included 389 NPC patients who received NACT from three institutions. Radiomics features were extracted from magnetic resonance imaging scans, while pathomics features were derived from histopathological images. A total of 2,667 radiomics features and 254 pathomics features were initially extracted. Feature selection involved intra-class correlation coefficient evaluation, Mann-Whitney U test, Spearman correlation analysis, and least absolute shrinkage and selection operator regression. The PSO-SVM model was constructed and validated using 10-fold cross-validation on the training set and further evaluated using an external validation set. Model performance was assessed using the area under the curve (AUC) of the receiver operating characteristic curve, calibration curves, and decision curve analysis.ResultsEight significant predictive features (five radiomics and three pathomics) were identified. The PSO-SVM radiopathomics model achieved superior performance compared to models based solely on radiomics or pathomics features. The AUCs for the PSO-SVM radiopathomics model were 0.917 (95% CI: 0.887–0.948) in internal validation and 0.814 (95% CI: 0.742–0.887) in external validation. Calibration curves demonstrated good agreement between predicted probabilities and actual outcomes. Decision curve analysis showed that the PSO-SVM radiopathomics model provided higher clinical net benefit over a wider range of risk thresholds compared to other models.ConclusionThe PSO-SVM radiopathomics model effectively integrates radiomics and pathomics features, offering enhanced predictive accuracy and clinical utility for assessing NACT efficacy in NPC. The multi-center approach and robust validation underscore its potential for personalized treatment planning, supporting improved clinical decision-making for NPC patients.

Prognostic value of nomogram model based on clinical risk factors and CT radiohistological features in hypertensive intracerebral hemorrhage.

To construct a nomogram model based on clinical risk factors and CT radiohistological features to predict the prognosis of hypertensive intracerebral hemorrhage (HICH). A total of 148 patients with HICH from April 2022 to July 2024 were retrospectively selected as the research subjects. According to the modified Rankin scale at the time of discharge, they were divided into good group (Rankin scale score 0-2) and bad group (Rankin scale score 3-6). To compare the clinical data and the changes of CT radiographic characteristics in patients with different prognosis. Relevant factors affecting the prognosis were analyzed, and nomogram model was established based on the influencing factors. The fitting degree, prediction efficiency and clinical net benefit of the nomogram model were evaluated by calibration curve, ROC curve and clinical decision curve (DCA). Compared with the good group, the hematoma volume in the poor group was significantly increased, the serum thromboxane 2(TXB2) and lysophosphatidic acid receptor 1(LPAR1) levels were significantly increased, and the energy balance related protein (Adropin) level was significantly decreased. The proportions of irregular shape, promiscuous sign, midline displacement, island sign and uneven density were all significantly increased (p < 0.05). In Logistic multivariate analysis, hematoma volume, Adropin, TXB2, LPAR1 and CT radiological features were all independent factors influencing the poor prognosis of HICH (p < 0.05). A nomogram prediction model was established based on the influencing factors. The calibration curve showed that the C-index was 0.820 (95% CI: 0.799-0.861), the goodness of fit test χ2 = 5.479, and p = 0.391 > 0.05, indicating a high degree of fitting. The ROC curve showed that the AUC was 0.896 (95% CI: 0.817-0.923), indicating that this model had high prediction ability. The DCA curve shows that the net benefit of the nomogram model is higher when the threshold probability is 0.1-0.9. The nomogram prediction model established based on hematoma volume, Adropin, TXB2, LPAR1 and other clinical risk factors as well as CT radiographic characteristics has high accuracy and prediction value in the diagnosis of poor prognosis in patients with HICH.

A Framework for Considering the Value of Race and Ethnicity in Estimating Disease Risk.

Accounting for race and ethnicity in estimating disease risk may improve the accuracy of predictions but may also encourage a racialized view of medicine. To present a decision analytic framework for considering the potential benefits of race-aware over race-unaware risk predictions, using cardiovascular disease, breast cancer, and lung cancer as case studies. Cross-sectional study. NHANES (National Health and Nutrition Examination Survey), 2011 to 2018, and NLST (National Lung Screening Trial), 2002 to 2004. U.S. adults. Starting with risk predictions from clinically recommended race-aware models, the researchers generated race-unaware predictions via statistical marginalization. They then estimated the utility gains of the race-aware over the race-unaware models, based on a simple utility function that assumes constant costs of screening and constant benefits of disease detection. The race-unaware predictions were substantially miscalibrated across racial and ethnic groups compared with the race-aware predictions as the benchmark. However, the clinical net benefit at the population level of race-aware predictions over race-unaware predictions was smaller than expected. This result stems from 2 empirical patterns: First, across all 3 diseases, 95% or more of individuals would receive the same decision regardless of whether race and ethnicity are included in risk models; second, for those who receive different decisions, the net benefit of screening or treatment is relatively small because these patients have disease risks close to the decision threshold (that is, the theoretical "point of indifference"). When used to inform rationing, race-aware models may have a more substantial net benefit. For illustrative purposes, the race-aware models were assumed to yield accurate estimates of risk given the input variables. The researchers used a simplified approach to generate race-unaware risk predictions from the race-aware models and a simple utility function to compare models. The analysis highlights the importance of foregrounding changes in decisions and utility when evaluating the potential benefit of using race and ethnicity to estimate disease risk. The Greenwall Foundation.

Predicting In-Hospital Mortality in Critical Orthopaedic Trauma Patients with Sepsis Using Machine Learning Models.

Purpose: This study aims to establish and validate machine learning-based models to predict death in hospital among critical orthopaedic trauma patients with sepsis or respiratory failure.Methods: This study collected 523 patients from the Medical Information Mart for Intensive Care database. All patients were randomly classified into a training cohort and a validation cohort. Six algorithms, including logistic regression (LR), extreme gradient boosting machine (eXGBM), support vector machine (SVM), random forest (RF), neural network (NN), and decision tree (DT), were used to develop and optimize models in the training cohort, and internal validation of these models were conducted in the validation cohort. Based on a comprehensive scoring system, which incorporated ten evaluation metrics, the optimal model was obtained with the highest scores. An artificial intelligence (AI) application was deployed based on the optimal model in the study.Results: The in-hospital mortality was 19.69%. Among all developed models, the eXGBM had the highest area under the curve (AUC) value (0.951, 95%CI: 0.934-0.967), and it also showed the highest accuracy (0.902), precise (0.893), recall (0.915), and F1 score (0.904). Based on the scoring system, the eXGBM had the highest score of 53, followed by the RF model (43) and the NN model (39). The scores for the LR, SVM, and DT were 22, 36, and 17, respectively. The decision curve analysis confirmed that both the eXGBM and RF models provided substantial clinical net benefits. However, the eXGBM model consistently outperformed the RF model across multiple evaluation metrics, establishing itself as the superior option for predictive modeling in this scenario, with the RF model as a strong secondary choice. The SHAP analysis revealed that SAPS II, age, respiratory rate, OASIS, and temperature were the most important five features contributing to the outcome.Conclusions: This study develops an artificial intelligence application to predict in-hospital mortality among critical orthopaedic trauma patients with sepsis or respiratory failure.

A Retrospective Study of Patiromer as Adjunct to Insulin Therapy for Acute Hyperkalemia in the Emergency Department.

To investigate the clinical utility of administering patiromer as an adjunct to insulin for potassium reduction in patients presenting to the emergency department (ED) with hyperkalemia. This retrospective cohort study used electronic health record data to identify adults treated with at least one intravenous dose of regular insulin for hyperkalemia within the ED. Patients who were administered patiromer within one hour before or after their insulin dose were categorized as the intervention group. Matching was performed at a 1:1 ratio. The primary outcome, mean change in potassium from baseline to the latest value within the 4-12 hour interval, was compared. Secondary outcomes included net clinical benefit, defined as the mean difference in the number of potassium-lowering interventions minus the change in potassium. The final analysis included 133 patients treated with patiromer plus insulin and 133 patients treated with insulin alone. Participants had a mean age of 71 years; 43% were female, 31% self-identified as Black, and 38% self-identified as Latinx. No significant changes were observed in potassium from baseline (mean levels 6.2 mEq/L in each group) to the 4-12 hour time frame (patiromer: -0.90 mEq/L, n=78 vs insulin-only: -0.98 mEq/L, n=81; p = 0.51). The calculated net clinical benefit of potassium reduction was -0.25 in favor of the patiromer plus insulin group; however, this difference did not reach statistical significance. In the subgroup of eGFR >30 mL/min, patiromer group received numerically less potassium-lowering interventions (0.63 vs 1.12, p = 0.057). In this study of patients with acute hyperkalemia in the ED setting, concurrent administration of patiromer did not result in more sustained potassium reduction compared to insulin alone in the overall cohort. The trend in favor of adjunct patiromer in the subgroup with adequate renal function warrants further investigation.

A nomogram model for predicting advanced liver fibrosis in patients with hepatitis B: A multicenter study.

Biopsy is the gold standard method for diagnosing liver fibrosis. FibroScan is a non-invasive method of diagnosing liver fibrosis, but it still faces some limitations. This study aimed to establish a nomogram model and identify patients at high risk of advanced liver fibrosis associated with hepatitis B infection. Data were collected from 375 patients with hepatitis B who underwent liver biopsy. Patients were divided randomly into the training (n = 263) and validation sets (n = 112). Their demographic and clinical characteristics were analyzed using the least absolute shrinkage and selection operator regression (LASSO). A nomogram model was established to predict the fibrosis stage, and its performance was assessed using the area under the receiver operating characteristic curve (AUC), calibration curve, and decision curve analysis (DCA) and was compared with other recognized models. In total, 209 patients with non-advanced fibrosis (S0-1) and 166 patients with advanced fibrosis (S ≥ 2) were included. Hyaluronic acid (HA), laminin, total cholesterol (TC), platelet, and age were entered into the nomogram model based on the LASSO analysis. The nomogram model for predicting advanced fibrosis exhibited a relatively high AUC in the training set. Compared with FIB4 and APRI, the nomogram model showed a better agreement between the actual status and predicted status based on the calibration curve. The nomogram model showed an AUC similar to FibroScan in the validation cohort, and showed high clinical net benefits in the training and validation sets. Our nomogram model can help identify patients with hepatitis B and advanced liver fibrosis.

Novel study on the prediction of BI-RADS 4A positive lesions in mammography using deep learning technology and clinical factors.

The classification of Breast Imaging Reporting and Data System (BI-RADS) category 4A lesions in mammography is complicated by subjective interpretations and unclear criteria, which can lead to potential misclassifications and unnecessary biopsies. Thus, more accurate assessment methods need to be developed. This study aimed to improve the classification prediction of BI-RADS 4A positive lesions in mammography by combining deep learning (DL) technology with relevant clinical factors. A retrospective analysis of 590 patients diagnosed with BI-RADS 4A at Shenzhen People's Hospital and Shenzhen Luohu People's Hospital was conducted, and a multi-faceted approach was employed to construct a robust predictive model. The patients were divided into training, validation, and external validation sets. The classification results from a DL system applied to mammography were recorded, and data on relevant clinical factors were collected. Univariate and multivariate logistic regression analyses were performed to identify the independent predictive factors. A predictive model and nomogram integrating these factors were developed. Assessment metrics, such as the areas under the curve (AUCs), calibration curves, and a decision curve analysis (DCA), were employed to evaluate the diagnostic performance, calibration, and clinical net benefit of the model. External validation was conducted to assess the generalization ability of the model. Four independent predictive factors (i.e., age, nipple discharge, ultrasound BI-RADS assessment, and DL system classification results) were identified and included in the predictive model. The model showed commendable diagnostic performance with AUC values of 0.85, 0.82, and 0.84 for the training, validation, and external validation sets, respectively. There were no statistically significant differences in the AUCs of the predictive model between the training set, and the internal and external validation sets (P=0.543 and 0.842, respectively). The calibration curves showed excellent calibration in the training, validation, and external validation sets, indicating a minimal deviation between the predicted and actual positive risk probabilities (P=0.906, 0.890, and 0.769, respectively). The DCA results illustrated the clinical net benefit of the model for risk thresholds greater than 0.15 and less than 0.70 in both the internal validation and external validation sets. Our predictive model, which incorporated age, nipple discharge, ultrasound BI-RADS assessment, and DL system classification results, emerged as a powerful tool for accurately predicting BI-RADS 4A positive lesions. Its application holds significant promise in helping radiologists enhance diagnostic precision and reduce unnecessary biopsies in BI-RADS 4A positive lesion cases.

Optimization of dual-energy CT enterography parameters for the assessment of Crohn's disease activity: a retrospective study.

Discuss the selection of imaging parameters in dual-energy CT enterography (DECTE) assessment for patients with Crohn's disease (CD) at different activity levels. This study analyzed data from 55 CD patients who had DECTE and endoscopy from 2020 to 2022. Patients were divided into moderate-severe (Crohn's Disease Endoscopic Index of Severity (CDEIS) ≥ 10) and remission-mild (CDEIS < 10) groups. Imaging indicators such as intestinal wall thickness, iodine concentration (IC), and lesion wall enhancement (ΔHu) were compared, and their correlations with CDEIS scores were analyzed. The receiver operating characteristic (ROC) curve was used to evaluate DECTE's effectiveness in assessing CD activity with multiparameter versus single-parameter methods. Decision curve analysis (DCA) and calibration curve analysis (CCA) were applied to assess patient benefits from the multiparameter assessment model. IC, ΔHu, wall thickness, and comb sign varied significantly between groups (P < 0.05). In the moderate-severe group, ΔHu correlated more strongly with CDEIS than in the remission-mild group. Lesion IC also highly correlated with CDEIS in both groups. DECTE, using multiparameter assessment, achieved an AUC of 0.933(95% CI 0.982-0.999, sensitivity was 0.906, specificity was 0.870) for diagnosing moderate to severe CD activity, surpassing single parameters assessment. CCA and DCA showed that multiparameter assessment had good calibration and net clinical benefits. Compared to moderate-severe CD, the assessment of disease activity in patients with mild-moderate CD was more important using DECTE. A multiparameter assessment can enhance the diagnostic efficacy of DECT for patients with CD.

Prediction of acute pancreatitis severity based on early CT radiomics

BackgroundThis study aims to develop and validate an integrated predictive model combining CT radiomics and clinical parameters for early assessment of acute pancreatitis severity.MethodsA retrospective cohort of 246 patients with acute pancreatitis was analyzed, with a 70%-30% split for training and validation groups. CT image segmentation was performed using ITK-SNAP, followed by the extraction of radiomics features. The stability of the radiomics features was assessed through inter-observer Intraclass Correlation Coefficient analysis. Feature selection was carried out using univariate analysis and least absolute shrinkage and selection operator (LASSO) regression with 10-fold cross-validation. A radiomics model was constructed through logistic regression to compute the radiomics score. Concurrently, univariate and multivariate logistic regression were employed to identify independent clinical risk factors for the clinical model. The radiomics score and clinical variables were integrated into a combined model, which was visualized with a nomogram. Model performance and net clinical benefit were evaluated through the area under the receiver operating characteristic curve (AUC), the DeLong test, and decision curve analysis.ResultsA total of 913 radiomics features demonstrated satisfactory consistency. Eight features were selected for the radiomics model. Serum calcium, C-reactive protein, and white blood cell count were identified as independent clinical predictors. The AUC of the radiomics model was 0.871 (95% CI, 0.793–0.949) in the training cohort and 0.859 (95% CI, 0.751–0.967) in the validation cohort. The clinical model achieved AUCs of 0.833 (95% CI, 0.756–0.910) and 0.810 (95% CI, 0.692–0.929) for the training and validation cohorts, respectively. The combined model outperformed both the radiomics and clinical models, with an AUC of 0.905 (95% CI, 0.837–0.973) in the training cohort and 0.908 (95% CI, 0.824–0.992) in the validation cohort. The DeLong test confirmed superior predictive performance of the combined model over both the radiomics and clinical models in the training cohort, and over the clinical model in the validation cohort. Decision curve analysis further demonstrated that the combined model provided greater net clinical benefit than the radiomics or clinical models alone.ConclusionThe clinical-radiomics model offers a novel tool for the early prediction of acute pancreatitis severity, providing valuable support for clinical decision-making.

Influencing Factors of Urinary Tract Stones Complicated by Urinary Tract Infections and the Construction of a Column Chart Prediction Model.

Objective: To analyze the influencing factors of urinary tract stones complicated by urinary tract infections and construct a column chart prediction model. Patients and Methods: From July 2020 to October 2023, 345 patients with urinary tract stones admitted to our hospital were collected as the training set, they were separated into an infection group of 51 cases and a non-infection group of 294 cases on the basis of the presence or absence of concurrent urinary tract infections; 192 patients with urinary tract stones were used as the testing set and were divided into an infection group of 26 cases and a non-infection group of 166 cases on the basis of the presence or absence of concurrent urinary tract infections. Data such as gender, age, and procalcitonin (PCT) were recorded. Multi-variable logistic regression analysis was applied to screen predictive factors, R4.0.2 software was applied to construct a column chart model, the calibration curve and Receiver Operating Characteristic (ROC) curve were applied to evaluate the discrimination and calibration of the column chart model; decision curve analysis curve was applied to evaluate the predictive performance of column chart models. Results: The proportions of female, diabetes mellitus, indwelling time of urinary catheter ≥7 days, the PCT, and urine pH in the infected group were greater than those in the non-infected group (p < 0.05). Female, diabetes mellitus, catheter retention time ≥7 days, high PCT, and high urine pH were independent risk factors for urinary calculi complicated with urinary tract infection (p < 0.05). Training set: C-index was 0.913, Area Under Curve (AUC) was 0.943 [95% Confidence Interval (CI) = 0.912-0.973], sensitivity was 86.36%, and specificity was 89.81%, testing set: C-index was 0.905, AUC was 0.959 (95% CI = 0.928-0.989), sensitivity was 84.65%, and specificity was 95.84%, indicating good discriminability of the line graph model; Hosmer-Lemeshow test showed χ2 = 2.843, 2.894, p = 0.944, 0.941, the calibration curve approached the ideal curve, and the line graph model had good calibration. When the risk threshold for urinary tract stones complicated by urinary tract infections was between 0.08 and 0.86, this column chart model provided clinical net benefits. Conclusion: The column chart prediction model for urinary tract stones complicated by urinary tract infections constructed in this study has high predictive efficiency and clinical practical value, and can provide reference for medical staff.

Risk prediction model for psoriatic arthritis: NHANES data and multi-algorithm approach.

To develop a simplified predictive model for identifying psoriatic arthritis (PsA) in psoriasis patients. Data from the National Health and Nutrition Examination Survey (NHANES) database were analyzed, including patients with psoriasis without arthritis (PsC) or PsA. The least absolute shrinkage and selection operator, Boruta algorithm, random forest, and stepwise regression were employed to select key variables from 38 potential predictors. Logistic regression models were constructed for each combination of selected variables and evaluated using receiver operating characteristic (ROC) curves, precision-recall (PR) curves, calibration plots, Brier scores, and decision curve analysis (DCA). The study included 587 patients with psoriasis, 238 of whom had PsA. The variable combinations proposed by the Boruta algorithm exhibited the best overall performance. Key predictors in the Borutamodel included age, fasting glucose, education level, thyroid disease, hypertension, and chronic bronchitis. This model achieved area under the curve (AUC) of 0.781 (95% CI, 0.737-0.826) for the training set and 0.780 (95% CI, 0.712-0.848) for the testing set in the ROC curve analyses. The AUC values in the PR curves were 0.687 (95% CI, 0.611-0.757) and 0.653 (95% CI, 0.535-0.770), respectively. The Brier scores of 0.186 and 0.191 for the testing and training sets indicated a good fit, further supported by the calibration curves. DCA showed a net clinical benefit for decision thresholds ranging from 0.2 to 0.8 in both datasets. The Borutamodel represents a promising tool for early risk assessment of PsA. Key Points • National Database Utilization: This study leverages the NHANES database to predict psoriatic arthritis risk, addressing previous limitations tied to regional or ethnic constraints. • Comprehensive Variable Analyses: The research examines 38 variables, including demographics, health conditions, laboratory results, and lifestyle factors, using four distinct screening methods and thorough evaluations of model performance. • Innovative Risk Model: The study introduces a novel risk assessment model that integrates age, fasting glucose, education, and comorbidities including hypertension, thyroid disease, and chronic bronchitis, thus moving beyond traditional focus on skin lesions and joint symptoms.

Serum homocysteine and left ventricular hypertrophy in adults with chronic kidney disease: A case–control study

Hyperhomocysteinemia (serum homocysteine concentration &gt; 15 μmol/L) is of high prevalence in chronic kidney disease (CKD). And myocardial hypertrophy is a common complication of CKD. Given that both hyperhomocysteinemia and cardiac hypertrophy have an association with CKD, we hypothesized that high level of plasma homocysteine (Hcy) is associated with a higher prevalence of ventricular hypertrophy(LVH) in adults with CKD. The registration number of the case-control study is ChiCTR2200064834. The information of inpatients with CKD including Echocardiograms and analysis of plasma Hcy concentrations were collected. We performed linear and logistic regression to investigate the association of plasma Hcy with left ventricular hypertrophy (LVH) (LVMI ≥ 95th percentile), adjusted for levels of hemoglobin, ferritin, cystatin C and β-adrenergic blocker therapy. Further, a stratified analysis of the relationship between plasma Hcy and LVH was carried out according to eGFR. The case records for 1068 inpatients with CKD were collected. After data soring and case-control matching, there were 374 samples screened for statistical analysis. Univariate logistic regression indicated a high level of serum Hcy had an association with LVH (OR, 1.16; 95% CI, 1.11–1.20). Finally, multivariable logistic regression suggested that hyperhomocysteinemia was independently associated with LVH (OR, 1.14; 95% CI, 1.10–1.19) after adjustment for hemoglobin, ferritin, cystatin C, and β-adrenergic receptor blocker therapy. We constructed a predicting model including the variable of Hcy for cardiac hypertrophy in CKD. The model had an area under the ROC curve (AUC) of 0.86 (95% CI: 0.82–0.89, P &lt; .001). The decision curve analysis (DCA) showed a superior net clinical benefit of model with Hcy over model without Hcy. Elevated level of serum Hcy is closely associated with LVH in adults with CKD.

Development and validation of a nomogram for obesity and related factors to detect gastric precancerous lesions in the Chinese population: a retrospective cohort study.

The purpose of this study was to construct a nomogram to identify patients at high risk of gastric precancerous lesions (GPLs). This identification will facilitate early diagnosis and treatment and ultimately reduce the incidence and mortality of gastric cancer. In this single-center retrospective cohort study, 563 participants were divided into a gastric precancerous lesion (GPL) group (n=322) and a non-atrophic gastritis (NAG) group (n=241) based on gastroscopy and pathology results. Laboratory data and demographic data were collected. A derivation cohort (n=395) was used to identify the factors associated with GPLs to develop a predictive model. Then, internal validation was performed (n=168). We used the area under the receiver operating characteristic curve (AUC) to determine the discriminative ability of the predictive model; we constructed a calibration plot to evaluate the accuracy of the predictive model; and we performed decision curve analysis (DCA) to assess the clinical practicability predictive model. Four -predictors (i.e., age, body mass index, smoking status, and -triglycerides) were included in the predictive model. The AUC values of this predictive model were 0.715 (95% CI: 0.665-0.765) and 0.717 (95% CI: 0.640-0.795) in the derivation and internal validation cohorts, respectively. These values indicated that the predictive model had good discrimination ability. The calibration plots and DCA suggested that the predictive model had good accuracy and clinical net benefit. The Hosmer-Lemeshow test results in the derivation and validation cohorts for this predictive model were 0.774 and 0.468, respectively. The nomogram constructed herein demonstrated good performance in terms of predicting the risk of GPLs. This nomogram can be beneficial for the early detection of patients at high risk of GPLs, thus facilitating early treatment and ultimately reducing the incidence and mortality of gastric cancer.

Machine learning for stroke in heart failure with reduced ejection fraction but without atrial fibrillation: A post‐hoc analysis of the WARCEF trial

AbstractBackgroundThe prediction of ischaemic stroke in patients with heart failure with reduced ejection fraction (HFrEF) but without atrial fibrillation (AF) remains challenging. Our aim was to evaluate the performance of machine learning (ML) in identifying the development of ischaemic stroke in this population.MethodsWe performed a post‐hoc analysis of the WARCEF trial, only including patients without a history of AF. We evaluated the performance of 9 ML models for identifying incident stroke using metrics including area under the curve (AUC) and decision curve analysis. The importance of each feature used in the model was ranked by SAPley Additive exPlanations (SHAP) values.ResultsWe included 2213 patients with HFrEF but without AF (mean age 58 ± 11 years; 80% male). Of these, 74 (3.3%) had an ischaemic stroke in sinus rhythm during a mean follow‐up of 3.3 ± 1.8 years. Out of the 29 patient‐demographics variables, 12 were selected for the ML training. Almost all ML models demonstrated high AUC values, outperforming the CHA2DS2‐VASc score (AUC: 0.643, 95% confidence interval [CI]: 0.512–0.767). The Support Vector Machine (SVM) and XGBoost models achieved the highest AUCs, with 0.874 (95% CI: 0.769–0.959) and 0.873 (95% CI: 0.783–0.953), respectively. The SVM and LightGBM consistently provided significant net clinical benefits. Key features consistently identified across these models were creatinine clearance (CrCl), blood urea nitrogen (BUN) and warfarin use.ConclusionsMachine‐learning models can be useful in identifying incident ischaemic strokes in patients with HFrEF but without AF. CrCl, BUN and warfarin use were the key features.

Construction and validation of a nomogram prediction model for the risk of new-onset atrial fibrillation following percutaneous coronary intervention in acute myocardial infarction patients

ObjectiveThe objective of this study was to investigate risk factors for new-onset atrial fibrillation (NOAF) post-percutaneous coronary intervention (PCI) in patients with acute myocardial infarction (AMI), aiming to develop a predictive nomogram for NOAF risk.MethodsA retrospective cohort study involving 397 AMI patients who underwent PCI at a tertiary hospital in Anhui, China, from January 2021 to July 2022 was performed. Patients were divided into NOAF (n = 63) and non-NOAF (n = 334) groups based on post-PCI outcomes. Clinical data were extracted from the hospital information system (HIS) and analyzed using univariate and multivariate logistic regression to identify independent risk factors. A nomogram was generated utilizing R software (version 3.6.1), with its performance evaluated through receiver operating characteristic (ROC) curves, calibration curves, decision curve analysis (DCA), and Bootstrap resampling.ResultsIndependent risk factors for NOAF included age, left atrial diameter (LAD), Gensini score, N-terminal pro-B-type natriuretic peptide (NT-proBNP), alanine transaminase (ALT), low-density lipoprotein cholesterol (LDL-C), left ventricular end-systolic diameter (LVESD), and ventricular rate (P < 0.05). The nomogram’s ROC curve demonstrated an area under the curve (AUC) of 0.925 (95% CI: 0.887–0.963), supported by a Bootstrap-verified AUC of 0.924 (95% CI: 0.883–0.954), reflecting strong discriminative capability. The calibration curve indicated a mean absolute error (MAE) of 0.031 and 0.017 prior to and following Bootstrap verification, respectively, signifying robust calibration. The DCA curve illustrated that the nomogram offered optimal clinical net benefit for patients with a threshold probability of NOAF ranging from 0.01 to 0.99.ConclusionThe nomogram developed from independent risk factors for NOAF exhibits significant predictive accuracy and clinical relevance for evaluating the risk of NOAF in AMI patients following PCI, thereby enabling the identification of high-risk individuals for targeted interventions.

Abstract 4143270: PREDICTIVE MODELS AID PHYSICIAN PROGNOSTICATION: A SECONDARY ANALYSIS EVALUATING INTEGRATED MODEL AND PHYSICIAN PROGNOSTIC ESTIMATES IN PATIENTS WITH HEART FAILURE WITH REDUCED EJECTION FRACTION

Background: In recent studies from a multicenter Canadian cohort of outpatients with heart failure (HF), we found that model predictions were significantly more accurate than HF cardiologists. In this study, trying to mimic practice, we evaluated the additional predictive value and clinical impact of model predictions to refine physician estimated risk of 1-year mortality by combining model and physician estimates. Methods: We included consented consecutive HF outpatients (LVEF &lt;40%) followed at 11 HF clinics in Canada. HF cardiologists estimated patient 1-year mortality using their clinical judgment. We calculated model predicted mortality using the Seattle HF Model (SHFM). We followed patients for at least a year to record mortality (or urgent heart transplant or ventricular assist device implant as mortality-equivalent events). Using random forest survival model and cross-validation, we compared the performance SHFM and the HF cardiologist alone, and the integrated HF cardiologist and the SHFM predictions by evaluating model discrimination (c-statistic), calibration (observed vs predicted event rate), risk reclassification and clinical net benefit analyses. Results: Among 1,643 HF patients, 1-year event rate was 9% (95%CI 8%-11%). The SHFM had the adequate discrimination (c-statistic 0.76) and excellent calibration while cardiologists showed adequate discrimination (c-statistic 0.75) and poor calibration with significant risk overestimation ( Figure 1 ). When the SHFM estimates were added physician predictions, discrimination significantly improved (0.82, 95%CI 0.78-0.86) with excellent calibration. By risk reclassification analysis, among patients with events, HF cardiologist better reclassified 44% than the SHFM or the integrated model. Among patients without event, however, HF cardiologists worse risk-classified 52% in comparison to SHFM and 71% to the integrated model. By net clinical benefit analysis ( Figure 2 ), when the decision to treat involves patients with 1-year mortality of &gt;5%, SHFM predictions would lead to higher benefit than guiding care by physician judgement. Integrating model and HF cardiologist predictions led to minimally increased benefit in comparison to SHFM alone. Conclusions: Integrating prediction from the SHFM to physician judgment or using the SHFM alone showed superior accuracy than HF cardiologist predictions, proving that model-informed care may provide more accurate prognostic information to tailor clinical decision making.

Machine learning model for early prediction of survival in gallbladder adenocarcinoma: A comparison study

The prognosis for gallbladder adenocarcinoma (GBAC), a highly malignant cancer, is not good. In order to facilitate individualized risk stratification and improve clinical decision-making, this study set out to create and validate a machine learning model that could accurately predict early survival outcomes in GBAC patients. Five models—RSF, Cox regression, GBM, XGBoost, and Deepsurv—were compared using data from the SEER database (2010–2020). The dataset was divided into training (70 %) and validation (30 %) sets, and the C-index, ROC curves, calibration curves, and decision curve analysis (DCA) were used to assess the model's performance. At 1, 2, and 3-year survival intervals, the RSF model performed better than the others in terms of calibration, discrimination, and clinical net benefit. The most important predictor of survival, according to SHAP analysis, is AJCC stage. Patients were divided into high, medium, and low-risk groups according to RSF-derived risk scores, which revealed notable variations in survival results. These results demonstrate the RSF model's potential as an early survival prediction tool for GBAC patients, which could enhance individualized treatment and decision-making.

Development and validation of a novel comorbidity score specific for prostate cancer patients treated with robotic platform and its implication on DaVinci single-port system.

To develop and validate a novel Comorbidity score for Robotic Surgery (CRS) in predicting severe complications after robot-assisted radical prostatectomy (RARP). Furthermore, we investigated the impact of the surgical platform (Multi-Port - MP vs Single-Port - SP) according to this score. We included 2085 ("development cohort") and 595 ("validation cohort") patients undergoing RARP at two tertiary referral centers between 2014 and March 2024 in a retrospective study. Statistical analyses included validation of the Charlson Comorbidity Index (CCI) to predict 30-day severe complications (Clavien-Dindo ≥ 3a), development and external validation of CRS using calibration plots and decision curve analysis. Lastly, locally weighted scatterplot smoothing (LOWESS) analysis was used to graphically explore the impact of the robotic platform according to novel CRS. CCI exhibited limited predictive ability for severe complications (60% in the validation cohort). In multivariable logistic regression analyses testing the correlation between each condition included in CCI and severe complications, diabetes and myocardial infarction resulted as independent predictors (OR 1.75 [95%CI 1.05-2.82]; OR 1.92 [95%CI 1.26-2.88]) and were subsequently fitted into a multivariable logistic model including age, previous abdominal surgery and obesity (BMI > 30). The resulting predictive model demonstrated superior discrimination and clinical net benefit in predicting severe complications compared to CCI (AUC 64 vs 60%). At LOWESS analysis, SP platform was associated with lower risk of severe complications as CRS increased compared to MP system. The validated CRS showed better accuracy compared to CCI in predicting severe complications after RARP. Additionally, the use of SP robotic platform may reduce the risk of severe complications in highly comorbid patients according to CRS.

Construction of OSA-related hypertension prediction model based on nomogram

Objective:This study aimed to construct a risk prediction model for obstructive sleep apnea（OSA） related hypertension based on the nomogram, and to explore the independent risk factors for OSA-related hypertension, so as to provide reference for clinical treatment decision-making. Methods:The clinical data of OSA patients diagnosed by polysomnography from October 2019 to December 2021 were collected retrospectively and randomly divided into training sets and validation sets. A total of 1 493 OSA patients with 27 variables were included. The least absolute shrinkage and selection operator（Lasso） logistic regression model was used to select potentially relevant features and establish a nomogram for OSA-related hypertension.The performance and clinical benefits of this nomogram were verified in terms of discrimination, calibration ability and clinical net benefit. Results:Multivariate logistic regression showed that body mass index（BMI）, family history of hypertension, lowest oxygen saturation（LSaO2）, age and cumulative percentage of total sleep time with oxygen saturation below 90% were independent risk factors for OSA-related hypertension. Lasso logistic regression identified BMI, family history of hypertension, LSaO2 and age as predictive factors for inclusion in the nomogram. The nomogram provided a favorable discrimination, with a C-indexes of 0.835（95% confidence interval[CI ]0.806-0.863） 0.865（95%CI 0.829-0.900） for the training and validation cohort, respectively, and well calibrated. The clinical decision curve analysis displayed that the nomogram was clinically useful. Conclusion:Compared with cumulative percentage of total sleep time with blood oxygen saturation below 90%, LSaO2 may have a greater impact on the incidence of OSA-related hypertension, and the effects of different times and degrees of hypoxia on OSA-related hypertension should be further explored in the future. Apnea hypopnea index involvement is weak in predicting OSA-related hypertension, and the blood oxygen index may be a better predictor variable. Furthermore, we established a risk prediction model for OSA-related hypertension patients using nomogram, and demonstrated that this prediction model was helpful to identify high-risk OSA-related hypertension patients. This model can provide early and individualized diagnosis and treatment plans, protect patients from the serious.