Sequential Data–Based Patient Similarity Framework for Patient Outcome Prediction: Algorithm Development

Sequential information in electronic medical records is valuable and helpful for patient outcome prediction but is rarely used for patient similarity measurement because of its unevenness, irregularity, and heterogeneity. We aimed to develop a patient similarity framework for patient outcome predict...

Full description

Saved in:
Bibliographic Details
Published in:Journal of medical Internet research Vol. 24; no. 1; p. e30720
Main Authors: Wang, Ni, Wang, Muyu, Zhou, Yang, Liu, Honglei, Wei, Lan, Fei, Xiaolu, Chen, Hui
Format: Journal Article
Language:English
Published: Canada Journal of Medical Internet Research 06.01.2022
Gunther Eysenbach MD MPH, Associate Professor
JMIR Publications
Subjects:
Algorithms
Analysis
Cardiac patients
Classifiers
Clinical outcomes
Cluster Analysis
Computerized medical records
Cross-Sectional Studies
Data
Decomposition
Deep learning
Diabetes
Disease
Dynamic programming
Electronic Health Records
Heart attack
Heart attacks
Hospitalization
Hospitals
Humans
Information
Information technology
Laboratories
Measurement
Medical records
Mortality
Multimedia
Myocardial infarction
Neural networks
Neural Networks, Computer
Original Paper
Patient admissions
Patient outcomes
Patient Readmission
Patients
Precision medicine
Prediction models
Predictions
Prognosis
Readmission
Recurrent
Sequences
Short term memory
Time
Time series
Trends
Type 2 diabetes
outcome prediction
deep learning
patient similarity
electronic medical records
natural language processing
health data
time series
informatics
machine learning
acute myocardial infarction
ISSN:1438-8871, 1439-4456, 1438-8871
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Sequential information in electronic medical records is valuable and helpful for patient outcome prediction but is rarely used for patient similarity measurement because of its unevenness, irregularity, and heterogeneity. We aimed to develop a patient similarity framework for patient outcome prediction that makes use of sequential and cross-sectional information in electronic medical record systems. Sequence similarity was calculated from timestamped event sequences using edit distance, and trend similarity was calculated from time series using dynamic time warping and Haar decomposition. We also extracted cross-sectional information, namely, demographic, laboratory test, and radiological report data, for additional similarity calculations. We validated the effectiveness of the framework by constructing k-nearest neighbors classifiers to predict mortality and readmission for acute myocardial infarction patients, using data from (1) a public data set and (2) a private data set, at 3 time points-at admission, on Day 7, and at discharge-to provide early warning patient outcomes. We also constructed state-of-the-art Euclidean-distance k-nearest neighbor, logistic regression, random forest, long short-term memory network, and recurrent neural network models, which were used for comparison. With all available information during a hospitalization episode, predictive models using the similarity model outperformed baseline models based on both public and private data sets. For mortality predictions, all models except for the logistic regression model showed improved performances over time. There were no such increasing trends in predictive performances for readmission predictions. The random forest and logistic regression models performed best for mortality and readmission predictions, respectively, when using information from the first week after admission. For patient outcome predictions, the patient similarity framework facilitated sequential similarity calculations for uneven electronic medical record data and helped improve predictive performance.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:1438-8871
1439-4456
1438-8871
DOI:10.2196/30720