相关实验视频
Updated: Jun 23, 2026
10:36
Probiotic Studies in Neonatal Mice Using Gavage
Published on: January 27, 2019
18.3K
机器学习提高了乳酸在住院婴儿的预测效用
medRxiv : the preprint server for health sciences
|April 16, 2025
概括
机器学习准确地预测了高乳糖血症婴儿能量代谢 (IEEM) 的先天性错误. 这种方法可以改善诊断,而不仅仅是乳酸水平,有助于临界预后.
科学领域:
- 生物医学信息学 生物医学信息学
- 新生儿科学 新生儿科学
- 代谢障碍 代谢障碍 代谢障碍
背景情况:
- 超乳酸血是住院婴儿经常出现的病症.
- 确定超乳糖血症的原因对于适当的治疗和预后至关重要.
- 能量代谢的先天性错误 (IEEM) 是新生儿高乳糖血症的关键差异诊断.
研究的目的:
- 开发和验证一种机器学习模型,用于预测患有高乳糖血症的住院婴儿的IEEM.
- 为了比较机器学习模型的诊断实用性,仅用乳酸水平.
主要方法:
- 对0-90天龄的婴儿进行回顾性队列研究,乳酸水平>=5mmol/L.
- 机器学习模型 (随机森林,XGBoost) 使用临床和实验室数据进行训练和验证,包括血氨基酸和甲酸水平.
- 模型性能使用接收器操作特征曲线 (AUC-ROC) 下的面积来评估.
主要成果:
- 机器学习模型在预测IEEM时达到0.81的AUC-ROC,显著超过单独的乳酸 (AUC-ROC0.56).
- 与其他原因相比,患有IEEM的婴儿的乳酸盐水平明显高 (中位数为12.6 mmol/L).
- 总的来说,死亡率很高 (30%),在IEEM (51%) 的婴儿中特别高.
结论:
- 机器学习证明了在新生儿高乳糖血症中识别IEEM的高诊断实用性.
- 这种方法促进了复杂数据的计算机辅助解释,使得诊断更快,更准确.
- 早期和准确的IEEM诊断对于改善高乳糖血症婴儿的结果至关重要.
相关概念视频
01:30Development of Human Microbiota
The human microbiota begins developing at birth and undergoes continual change as we age. Infancy marks a critical period of microbial sensitivity, offering a “window of opportunity” during which beneficial microbes help mature the immune system. By age three, children typically develop a more stable and diverse microbial community. Newborns acquire microbes from their immediate environment; vaginal delivery favors maternal vaginal microbes, while cesarean births favor microbes from the skin...
01:28Development of the Oral Microbiota
The establishment of the oral microbiome begins before birth, challenging the long-held belief that the fetal oral cavity is sterile. The presence of oral microbes such as Streptococcus and Fusobacterium in amniotic fluid suggests that microbial exposure may occur in utero, potentially through translocation from the maternal oral or gastrointestinal tract. This early colonization primes the neonatal immune system and sets the stage for subsequent microbial succession. Maternal health,...