Interpolation of Missing Physiological Data of ICU Patients Based on Deep Embedded Clustering

doi:10.12068/j.issn.1005-3026.2022.05.005

Abstract

Abstract: The data in electronic medical records are often missing， significantly affecting the analysis results. The ICU(intensive care unit)patients’ data in MIMIC database were analyzed for missing value interpolation， and the dataset consists of 23 groups of commonly used clinical physiological variables and 5260 samples without missing values. A K-nearest neighbor interpolation method was proposed based on deep embedded clustering. This method takes deep embedded clustering as the core， constructs the sample proximity matrix through multiple clustering， and then regards the average value of K-nearest neighbor of missing samples as the missing values. Compared with mean interpolation， median interpolation， a posteriori distribution estimation interpolation and conditional mean interpolation， the proposed method obtains higher similarity between the interpolated results and the original data， and better retains the differences between various samples.

Key words: intensive care unit(ICU); electronic medical record; missing value interpolation; deep embedded clustering; proximity matrix

CLC Number:

TP391.4

LI Jian-hua， ZHU Ze-yang， XU Li-sheng， SUN Guo-zhe. Interpolation of Missing Physiological Data of ICU Patients Based on Deep Embedded Clustering[J]. Journal of Northeastern University(Natural Science), 2022, 43(5): 639-645.

References

[1]Digiovine B，Chenoweth C，Watts C，et al.The attributable mortality and costs of primary nosocomial bloodstream infections in the intensive care unit［J］.American Journal of Respiratory and Critical Care Medicine，1999，160(3):976-981.
[2]侯明曙.重症监护病房的发展现状［J］.现代医药卫生，1995(2):40.(Hou Ming-shu.Development status of intensive care unit ［J］.Modern Medicine and Health，1995(2):40.)
[3]Idri A，Benhar H，Fernández-Alemán J L，et al.A systematic map of medical data preprocessing in knowledge discovery［J］.Computer Methods and Programs Biomedicine，2018，162:69-85.
[4]El-Rashidy N，EI-Sappagh S，Abuhmed T，et al.Intensive care unit mortality prediction:an improved patient-specific stacking ensemble model［J］.IEEE Access，2020，8:133541-133564.
[5]Macas M，Kuzilek J，Odstrcilik T，et al.Linear Bayes classification for mortality prediction［C］// 2012 Computing in Cardiology.Krakow:IEEE，2012:473-476.
[6]Kim S，Kim W，Park R W.A comparison of intensive care unit mortality prediction models through the use of data mining techniques［J］.Healthcare Informatics Research，2011，17(4):232-243.
[7]Liu J，Chen X，Fang Z，et al.ICU mortality prediction using modified cost-sensitive PCA and chaos PSO［C］// 2017 IEEE the 19th International Conference on e-Health Networking，Applications and Services(Healthcom).Dalian:IEEE，2017:1-6.
[8]Johnson A E，Pollard T J，Shen L，et al.MIMIC-III，a freely accessible critical care database［J］.Scientific Data，2016，3:160035.
[9]Pickering B W，Gajic O，Ahmed A，et al.Data utilization for medical decision making at the time of patient admission to ICU［J］.Critical Care Medicine，2013，41(6):1502-1510.
[10]Peng X，Feng J，Zhou J T，et al.Deep subspace clustering［J］.IEEE Transactions on Neural Networks and Learning Systems，2020，31(12):5509-5521.
[11]Vincent P，Larochelle H，Bengio Y，et al.Extracting and composing robust features with denoising autoencoders［C］//Proceedings of the 25th International Conference on Machine Learning.Helsinki，2008:1096-1103.
[12]Yang B，Fu X，Sidiropoulos N D，et al.Towards k-means-friendly spaces:simultaneous deep learning and clustering［C］//Proceedings of the 34th International Conference on Machine Learning.Sydney，2017:3861-3870.
[13]Masci J，Meier U，Ciresan D，et al.Stacked convolutional auto-encoders for hierarchical feature extraction［C］//Proceedings of the 21th International Conference on Artificial Neural Networks.Espoo，2011:52-59.
[14]Jain A K.Data clustering:50 years beyond k-means［J］.Pattern Recognition Letters，2010，31(8):651-666.
[15]Dasuki Y D，Sheng O C，Hee L M，et al.A hybrid k-means-GMM machine learning technique for turbomachinery condition monitoring［C］//MATEC Web of Conferences.Sabah，2019:06008.
[16]Fred A L N，Jain A K.Combining multiple clusterings using evidence accumulation［J］.IEEE Transactions on Pattern Analysis and Machine Intelligence，2005，27(6):835-850.