{"manuscript_title":"<b>Development and Validation of a Nomogram to Predict Survival in Pulmonary Fibrosis Patients Admitted to the</b><b> ICU</b>","abstract":"<b>Background</b><b>: </b>Pulmonary interstitial fibrosis, a common end-stage manifestation of interstitial pneumonia, is a chronic and progressive lung disease marked by inflammatory cell infiltration and extensive fibrosis in the lung interstitium. Prompt identification of high-risk patients is crucial for guiding clinical decisions and determining the optimal timing for lung transplantation. This study aimed to develop and validate a nomogram to predict the overall survival rate of patients with pulmonary fibrosis (PF) in the intensive care unit (ICU).<b>Methods:</b><b> </b>A total of 459 patients were included in this study from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. Within 24 hours of patient admission, 55 clinical indicators were collected. The least absolute selection and shrinkage operator (LASSO) was utilized with 10-fold cross-validation methods to select the optimal prognostic factors. Subsequently, multivariate COX regression analysis was performed to identify independent prognostic factors and construct the nomogram. Model performance and clinical utility were evaluated using the C-index, time-dependent receiver operating characteristic (ROC) curve, calibration curve, decision curve analysis (DCA), and Kaplan-Meier survival curve. Time-dependent ROC and DCA analyses were employed to compare the predictive value of the nomogram model with APSIII.<b>Results:</b><b> </b>The 30-day, 90-day, and 180-day OS rates of 459 patients with PF were 68.41%, 62.75%, and 56.43%, respectively.<b> </b>Indicators included in the nomogram were age, temperature, RDW, PaO<sub>2</sub>, and APSIII. The C-index of the training set was 0.688 and that of the validation set was 0.678; the time-dependent ROC, calibration curves and DCA of the two groups showed good discrimination and accuracy. When compared to APSIII, the nomogram model demonstrated greater accuracy in predicting survival rates.<b>Conclusion:</b><b> </b>This study has successfully developed and validated the initial predictive model that integrates five clinical features.  This model effectively forecasts short-term survival in ICU-admitted patients with PF for any reasons and promptly identifies high risk individuals.","keywords":["Nomogram","Pulmonary Fibrosis","Survival Rate"]}