使用常规医疗保健数据预测低资源新生儿单位早期发作的败血症风险:多变量统计和机器学习模型的开发和评估
Ed Lowther1, Nushrat Khan2,3, Mario Cortina-Borja3
1UCL Advanced Research Computing Centre, University College London, London, UK.
BMJ paediatrics open
|September 29, 2025
概括
机器学习模型显示出在低资源环境中诊断新生儿败血症的前景. 这些工具可以在血液培养无法获得时帮助分拣,改善新生儿的结果.
科学领域:
- 医疗信息学 医疗信息学
- 医疗保健中的机器学习
- 新生儿医学 新生儿医学
背景情况:
- 新生儿败血症是低资源地区疾病和死亡的重要原因.
- 准确和适合环境的诊断工具对于改善新生儿结果至关重要.
研究的目的:
- 评估机器学习算法,以诊断新生儿败血症在低资源环境中.
- 将机器学习模型的性能与新生儿败血症分类的传统方法进行比较.
主要方法:
- 利用Neotree数字健康工具对萨莉·穆加贝中央医院接受治疗的新生儿的数据.
- 使用临床医生诊断和血液培养结果建模了一个复合结果变量.
- 开发和评估了三个算法:LightGBM,后勤回归和k-最近邻居.
主要成果:
- 在18345名新生儿中确定了917例早期发病的新生儿败血症.
- 轻GBM实现了0.712的接收器操作特征曲线下的面积,超过了后勤回归 (0.687).
- k-最近邻近模型显示了可比性能 (0.699) 和可解释性.
结论:
- 机器学习在资源有限的环境中为新生儿败血症分类提供了潜在的优势.
- 机器学习模型提供了直观的预测,并且可以在没有归算的情况下处理缺失的数据.
- 虽然性能差异在统计学上并不显著,但ML显示出临床决策支持的前景.
相关概念视频
01:18Steps in Outbreak Investigation
492
In the ever-evolving field of public health, statistical analysis serves as a cornerstone for understanding and managing disease outbreaks. By leveraging various statistical tools, health professionals can predict potential outbreaks, analyze ongoing situations, and devise effective responses to mitigate impact. For that to happen, there are a few possible stages of the analysis:
492
01:22Healthcare Associated Infections II: Preventive Measures
3.6K
Essential infection prevention measures are based on the knowledge of the infection chain, the modes of transmission in healthcare settings, and the use of the best practices in all healthcare settings. Compulsory public reporting of healthcare-associated infection rates is needed to allow individuals and the community to make informed choices regarding selecting a healthcare facility.
The best practices for preventing healthcare-associated infections include hand hygiene, patient risk...
The best practices for preventing healthcare-associated infections include hand hygiene, patient risk...
3.6K