A comparison between discrete and continuous time Bayesian networks in learning from clinical time series data with irregularity

Autor: Arjen Hommersom, Erik Bischoff, Fabio Stella, Peter J. F. Lucas, Lonneke M. Boer, Manxia Liu
Přispěvatelé: Datamanagement & Biometrics, Department Computer Science, RS-Research Line Resilience (part of LIRS program), Liu, M, Stella, F, Hommersom, A, Lucas, P, Boer, L, Bischoff, E
Jazyk: angličtina
Rok vydání: 2019
Předmět:
Multivariate statistics
Computer science
Medicine (miscellaneous)
Strong prior
Irregular time-series data
computer.software_genre
DISEASE
Cohort Studies
Pulmonary Disease
Chronic Obstructive
03 medical and health sciences
All institutes and research themes of the Radboud University Medical Center
0302 clinical medicine
Artificial Intelligence
COPD EXACERBATIONS
Software Science
MANAGEMENT
Humans
Learning
COPD
Time series
Dynamic Bayesian network
030304 developmental biology
0303 health sciences
MEDICINE
INF/01 - INFORMATICA
Bayesian network
Bayes Theorem
Point evidence
Missing data
22/4 OA procedure
Dynamic Bayesian networks
Interval evidence
Discrete time and continuous time
Continuous-time Bayesian networks
Continuous-time Bayesian network
Inflammatory diseases Radboud Institute for Health Sciences [Radboudumc 5]
Snapshot (computer storage)
Data mining
computer
SYSTEM
030217 neurology & neurosurgery
Zdroj: Liu, M, Stella, F, Hommersom, A, Lucas, P J F, Boer, L & Bischoff, E 2019, ' A comparison between discrete and continuous time Bayesian networks in learning from clinical time series data with irregularity ', Artificial Intelligence in Medicine, vol. 95, pp. 104-117 . https://doi.org/10.1016/j.artmed.2018.10.002
Artificial intelligence in medicine, 95, 104-117. Elsevier
Artificial Intelligence in Medicine, 95, 104-117. Elsevier
Artificial Intelligence in Medicine, 95, 104-117
Artificial Intelligence in Medicine, 95, pp. 104-117
ISSN: 0933-3657
DOI: 10.1016/j.artmed.2018.10.002
Popis: Background:Recently, mobile devices, such as smartphones, have been introduced into healthcare research to substitute paper diaries as data-collection tools in the home environment. Such devices support collecting patient data at different time points over a long period, resulting in clinical time-series data with high temporal complexity, such as time irregularities. Analysis of such time series poses new challenges for machine-learning techniques. The clinical context for the research discussed in this paper is home monitoring in chronic obstructive pulmonary disease (COPD).Objective:The goal of the present research is to find out which properties of temporal Bayesian network models allow to cope best with irregularly spaced multivariate clinical time-series data.Methods:Two mainstream temporal Bayesian network models of multivariate clinical time series are studied: dynamic Bayesian networks, where the system is described as a snapshot at discrete time points, and continuous time Bayesian networks, where transitions between states are modeled in continuous time. Their capability of learning from clinical time series that vary in nature are extensively studied. In order to compare the two temporal Bayesian network types for regularly and irregularly spaced time-series data, three typical ways of observing time-series data were investigated: (1) regularly spaced in time with a fixed rate; (2) irregularly spaced and missing completely at random at discrete time points; (3) irregularly spaced and missing at random at discrete time points. In addition, similar experiments were carried out using real-world COPD patient data where observations are unevenly spaced.Results:For regularly spaced time series, the dynamic Bayesian network models outperform the continuous time Bayesian networks. Similarly, if the data is missing completely at random, discrete-time models outperform continuous time models in most situations. For more realistic settings where data is not missing completely at random, the situation is more complicated. In simulation experiments, both models perform similarly if there is strong prior knowledge available about the missing data distribution. Otherwise, continuous time Bayesian networks perform better. In experiments with unevenly spaced real-world data, we surprisingly found that a dynamic Bayesian network where time is ignored performs similar to a continuous time Bayesian network.Conclusion:The results confirm conventional wisdom that discrete-time Bayesian networks are appropriate when learning from regularly spaced clinical time series. Similarly, we found that time series where the missingness occurs completely at random, dynamic Bayesian networks are an appropriate choice. However, for complex clinical time-series data that motivated this research, the continuous-time models are at least competitive and sometimes better than their discrete-time counterparts. Furthermore, continuous-time models provide additional benefits of being able to provide more fine-grained predictions than discrete-time models, which will be of practical relevance in clinical applications.
Databáze: OpenAIRE
Externí odkaz: