arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2310.07799cs.LGcs.AI
收藏

通过桥接电子病历数据集间的数据分布偏移实现域不变临床表征学习

Domain-invariant Clinical Representation Learning by Bridging Data Distribution Shift across EMR Datasets

Affiliated Xuzhou Municipal Hospital of Xuzhou Medical University(徐州医科大学附属徐州医院);Peking University School and Hospital of Stomatology(北京大学口腔医学院);National Engineering Research Center for Software Engineering, Peking University(北京大学国家软件工程研究中心);Peking University Third Hospital(北京大学第三医院);University of Edinburgh(爱丁堡大学);Health Data Research UK(英国健康数据研究中心)

Zhongji Zhang, Yuhang Wang, Yinghao Zhu, Xinyu Ma, Yasha Wang, Junyi Gao, Liantao Ma, Wen Tang, Xiaoyun Zhang, Ling Wang

AI总结:

针对新兴疾病早期EMR数据有限、跨数据集特征错位问题,提出域不变表征学习方法,构建源目标数据集转换模型约束分布偏移,捕捉任务相关不变特征,实验表明其优于基线且收敛更快,可提升疾病预测准确性。

摘要:

新兴疾病因可用信息有限,给症状识别和及时临床干预带来挑战。有效的预后模型可协助医生做出准确诊断并制定个性化治疗方案,以预防不良结局。然而,在疾病出现的早期阶段,多种因素阻碍模型开发:数据收集有限、临床经验不足,以及隐私和伦理问题限制数据可用性并使准确标签分配复杂化。此外,来自不同疾病或来源的电子病历(EMR)数据通常存在显著的跨数据集特征错位,严重影响深度学习模型的有效性。我们提出一种域不变表征学习方法,构建源数据集与目标数据集间的转换模型。通过约束不同域生成特征的分布偏移,捕捉与下游任务相关的域不变特征,开发统一的域不变编码器,在各类任务域中实现更优的特征表征。多目标任务的实验结果表明,我们提出的模型优于竞争基线方法,且训练收敛更快,尤其在数据有限时。大量实验验证了我们的方法在为新兴大流行及其他疾病提供更准确预测方面的有效性。代码公开于https://github.com/wang1yuhang/domain_invariant_network。

英文摘要:

Emerging diseases present challenges in symptom recognition and timely clinical intervention due to limited available information. An effective prognostic model could assist physicians in making accurate diagnoses and designing personalized treatment plans to prevent adverse outcomes. However, in the early stages of disease emergence, several factors hamper model development: limited data collection, insufficient clinical experience, and privacy and ethical concerns restrict data availability and complicate accurate label assignment. Furthermore, Electronic Medical Record (EMR) data from different diseases or sources often exhibit significant cross-dataset feature misalignment, severely impacting the effectiveness of deep learning models. We present a domain-invariant representation learning method that constructs a transition model between source and target datasets. By constraining the distribution shift of features generated across different domains, we capture domain-invariant features specifically relevant to downstream tasks, developing a unified domain-invariant encoder that achieves better feature representation across various task domains. Experimental results across multiple target tasks demonstrate that our proposed model surpasses competing baseline methods and achieves faster training convergence, particularly when working with limited data. Extensive experiments validate our method's effectiveness in providing more accurate predictions for emerging pandemics and other diseases. Code is publicly available at https://github.com/wang1yuhang/domain_invariant_network.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑