Fig. 1 | Scientific Reports

Fig. 1

From: Development and validation of a predictive model for continuous renal replacement therapy in sepsis patients using the MIMIC-IV database

Fig. 1

(a) Flow chart of data screening (Flow diagram illustrating the selection process of ICU patients from the MIMIC-IV v3.0 database (2008–2022). A total of 94,458 ICU records were screened. Patients were included if they were admitted to the ICU for the first time during their first hospitalization and stayed for at least 24 h. Further exclusions were applied for end-stage renal disease, malignancy, or missing clinical data. After these criteria, 7361 septic patients were included in the final analysis). (b) Missing Data Proportions of Candidate Predictors for CRRT Prediction Modeling (Missing data proportions for all candidate clinical, demographic, and laboratory variables. Variables such as thrombin, D-dimer, and CRP exhibited high missingness, with rates exceeding 90%. In contrast, core demographic variables (e.g., age, gender) and key clinical indicators (e.g., blood pressure, creatinine, comorbidities) had minimal or no missing values).

Back to article page