开发和验证基于网络的,可解释的败血症和严重烧伤死亡率的预测模型
Shi-Qi Wang1,2, Kan Qiu3, Qi-Rui Zheng4
1Department of Burns and Plastic Surgery, Jinling Hospital, Jinling Clinical Medical College, Nanjing University of Chinese Medicine, Nanjing, Jiangsu, China.
Frontiers in cellular and infection microbiology
|September 3, 2025
概括
机器学习模型可以准确预测严重烧伤患者的败血症和死亡率,从而改善临床决策和患者的治疗结果. 这些先进的算法比传统的方法提供更好的预测.
科学领域:
- 医学研究
- 计算生物学
- 创伤手术
背景情况:
- 广泛的烧伤 (≥50%的总体表面积) 与高血和死亡率有关.
- 识别风险因素和开发预测模型对于管理这些严重病例至关重要.
研究的目的:
- 确定严重烧伤患者的败血症和死亡风险因素.
- 开发准确和可解释的机器学习模型来预测败血症和死亡率.
主要方法:
- 对237名严重烧伤患者的回顾性队列研究 (2012-2023年).
- 应用了十种机器学习算法 (例如,随机森林,梯度增强树) 来预测败血症和死亡率.
- 使用AUC,精度,回忆,精度,F1得分进行评估模型;与SOFA得分进行比较;使用SHAP进行解释.
主要成果:
- 主要的败血症预测指标:SOFA评分,新冲击,白蛋白,BUN,三度烧伤区域,TBSA烧伤,白细胞计数,吸入损伤.
- 主要死亡预测指标:ALT,SOFA得分,烧伤类型,新的冲击,三度烧伤区域,TBSA烧伤,败血症.
- 随机森林模型实现了高败血症预测 (AUC=0.977);梯度增强树模型在死亡率预测方面表现出色 (AUC=0.981).
结论:
- 机器学习模型 (RF,GBT) 可以有效预测严重烧伤患者的败血症和死亡率.
- SHAP分析提高了临床解释和早期干预的模型透明度.
- 开发了基于网络的计算器,以帮助临床决策和改善患者的结果.
相关概念视频
Burn Injuries
2.8K
Burn injuries occur when the skin and underlying tissues are damaged due to exposure to heat, electricity, chemicals, radiation, or friction. They can vary in severity, from minor superficial burns to severe deep burns that can be life-threatening.
The damage results in the death of skin cells, which can lead to a massive loss of fluid. Dehydration, electrolyte imbalance, and renal and circulatory failure follow, which can be fatal. Burn patients are treated with intravenous fluids to offset...
The damage results in the death of skin cells, which can lead to a massive loss of fluid. Dehydration, electrolyte imbalance, and renal and circulatory failure follow, which can be fatal. Burn patients are treated with intravenous fluids to offset...
2.8K
Data Validation
5.3K
Data validation is an essential part of a comprehensive assessment. Validation is confirming or verifying and opening the door to gathering more assessment data as it clarifies vague or unclear data. The process of checking and verifying the collected information is called data validation. The primary purpose of data validation is to ensure data is as free from error, bias, and misinterpretation as possible.
Nursing assessment guides are generally based on holistic models rather than medical...
Nursing assessment guides are generally based on holistic models rather than medical...
5.3K


