A biomarker combination for acute respiratory distress syndrome prognosis evaluation and use thereof
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 中国人民解放军总医院第八医学中心
- Filing Date
- 2025-12-15
- Publication Date
- 2026-06-12
AI Technical Summary
Existing diagnostic criteria and severity scoring systems for ARDS are insufficient to accurately predict the clinical outcomes of individual patients, failing to meet the needs for precise risk stratification and individualized treatment. Existing biomarker prediction models have also failed to effectively integrate multi-dimensional biological information.
Using chromosome structure maintenance protein 4 (SMC4) or indole derivatives (such as indoleacetic acid and 3-formylindole), Acinetobacter baumannii, pulmonary surfactant protein A (SFTPA1 and SFTPA2), tryptophan metabolism pathway-related genes and other biomarkers, combined with machine learning models, an ARDS prognostic assessment system was constructed to predict patients' risk of death by detecting the expression level or abundance of biomarkers.
It improved the sensitivity and specificity of predicting the risk of death from ARDS, increasing the AUC value from 0.7226 to 0.984, providing a more accurate prognostic assessment tool and having great clinical translational value.
Smart Images
Figure CN121577903B_ABST