典型文献
Federated Learning Based on Extremely Sparse Series Clinic Monitoring Data
文献摘要：
Decentralized machine learning frameworks, e.g., federated learning, are emerging to facilitate learning with medical data underprivacy protection. It is widely agreed that the establishment of an accurate and robust medical learning model requires a large number of continuoussynchronous monitoring data of patients from various types of monitoring facilities. However, the clinic monitoring data are usuallysparse and imbalanced with errors and time irregularity, leading to inaccurate risk prediction results. To address this issue, this paper designsa medical data resampling and balancing scheme for federated learning to eliminate model biases caused by sample imbalance and provideaccurate disease risk prediction on multi-center medical data. Experimental results on a real-world clinical database MIMIC-IV demonstratethat the proposed method can improve AUC (the area under the receiver operating characteristic) from 50.1％ to 62.8％, with a significant performanceimprovement of accuracy from 76.8％ to 82.2％, compared to a vanilla federated learning artificial neural network (ANN). Moreover,we increase the model's tolerance for missing data from 20％ to 50％ compared with a stand-alone baseline model.
文献关键词：
中图分类号：
作者姓名：
LU Feng;GU Lin;TIAN Xuehua;SONG Cheng;ZHOU Lun
作者机构：
School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan 430074, China;Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan 430074, China
文献出处：
引用格式：
[1]LU Feng;GU Lin;TIAN Xuehua;SONG Cheng;ZHOU Lun-.Federated Learning Based on Extremely Sparse Series Clinic Monitoring Data)[J].中兴通讯技术（英文版）,2022(03):27-34
A类：
underprivacy,continuoussynchronous,usuallysparse,designsa,provideaccurate,demonstratethat,performanceimprovement
B类：
Federated,Learning,Based,Extremely,Sparse,Series,Clinic,Monitoring,Data,Decentralized,machine,learning,frameworks,federated,emerging,facilitate,medical,protection,It,widely,agreed,establishment,robust,model,requires,large,number,monitoring,patients,from,various,types,facilities,However,imbalanced,errors,irregularity,leading,inaccurate,risk,prediction,results,To,address,this,issue,paper,resampling,balancing,scheme,eliminate,biases,caused,by,sample,disease,multi,center,Experimental,real,world,clinical,database,MIMIC,IV,proposed,method,area,receiver,operating,characteristic,significant,accuracy,compared,vanilla,artificial,neural,network,ANN,Moreover,increase,tolerance,missing,stand,alone,baseline
AB值：
0.567895
相似文献
机标中图分类号，由域田数据科技根据网络公开资料自动分析生成，仅供学习研究参考。
