可适应的图形神经网络设计,以支持临床事件预测的概括性
Amara Tariq1, Gurkiran Kaur2, Leon Su3
1Arizona Advanced AI (A3I) Hub, Mayo Clinic Arizona, United States.
Journal of biomedical informatics
|February 16, 2025
概括
基于图形的卷积神经网络 (GCNN) 为临床事件预测提供了可适应的解决方案,克服了电子医疗记录中的泛化挑战. 这些模型在外部验证方面表现出卓越的表现,改善了不同机构的预测.
科学领域:
- 医疗信息学 医疗信息学
- 机器学习 机器学习
- 人工智能的人工智能
背景情况:
- 使用电子医疗记录 (EHR) 的临床事件预测模型通常由于机构差异而在外部验证中失败.
- 现有的模型在医疗保健机构中与患者群体和医疗实践的变化作斗争.
研究的目的:
- 开发和评估可系统地适应的基于图形的卷积神经网络 (GCNN),以进行可靠的临床事件预测.
- 解决机器学习模型在不同临床环境中面临的泛化挑战.
主要方法:
- 提出了一种新的GCNN架构,旨在实现系统的适应性.
- 利用GCNNs中隐式使用图形编码数据,允许在不需要再培训的情况下进行培训后的适应.
- 集成的多式联运数据用于增强的预测能力.
主要成果:
- 可适应的GCNN模型在医院出院和死亡率预测的外部验证方面显著优于比较模型 (AUROC 0.70和0.91对比0.58和0.81).
- 与基线模型相比,GCNNs在预测计划外输血 (AUROC 0.70) 中表现优异,即使外部数据不完整.
- 这些模型有效地支持多式联运数据集成.
结论:
- 这项研究支持这样的假设,即设计好的GCNNs可以克服临床预测模型中的概括问题.
- 可适应的GCNN为在不同医疗机构中开发可靠的预测工具提供了有希望的方法.
更多相关视频
03:37Author Spotlight: Impact of Intergenic Interactions on Disease-Identifying Dark Biomarkers
Published on: March 1, 2024
624
10:02Event Related Potentials ERPs and other EEG Based Methods for Extracting Biomarkers of Brain Dysfunction: Examples from Pediatric Attention Deficit/Hyperactivity Disorder ADHD
Published on: March 12, 2020
15.5K
相关概念视频
Kaplan-Meier Approach
79
The Kaplan-Meier estimator is a non-parametric method used to estimate the survival function from time-to-event data. In medical research, it is frequently employed to measure the proportion of patients surviving for a certain period after treatment. This estimator is fundamental in analyzing time-to-event data, making it indispensable in clinical trials, epidemiological studies, and reliability engineering. By estimating survival probabilities, researchers can evaluate treatment effectiveness,...
79
Steps in Outbreak Investigation
102
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:
102
