Classification of patients using novel multivariate time series representations of physiological data

Patricia Ordóñez; Tom Armstrong; Tim Oates; Jim Fackler

doi:10.1109/ICMLA.2011.46

Classification of patients using novel multivariate time series representations of physiological data

Patricia Ordóñez, Tom Armstrong, Tim Oates, Jim Fackler

School of Medicine

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

5 Scopus citations

Abstract

In this paper we present two novel multivariate time series representations to classify physiological data of different lengths. The representations may be applied to any group of multivariate time series data that examine the state or health of an entity. Multivariate Bag-of-Patterns and Stacked Bags of-Patterns improve on their univariate counterpart, inspired by the bag-of-words model, by using multiple time series and analyzing the data in a multivariate fashion. We also borrow techniques from the natural language processing domain such as term frequency and inverse document frequency to improve classification accuracy. We introduce a technique named inverse frequency and present experimental results on classifying patients who have experienced acute episodes of hypotension.

Original language	English (US)
Title of host publication	Proceedings - 10th International Conference on Machine Learning and Applications, ICMLA 2011
Pages	172-179
Number of pages	8
DOIs	https://doi.org/10.1109/ICMLA.2011.46
State	Published - 2011
Event	10th International Conference on Machine Learning and Applications, ICMLA 2011 - Honolulu, HI, United States Duration: Dec 18 2011 → Dec 21 2011

Publication series

Name	Proceedings - 10th International Conference on Machine Learning and Applications, ICMLA 2011
Volume	2

Other

Other	10th International Conference on Machine Learning and Applications, ICMLA 2011
Country/Territory	United States
City	Honolulu, HI
Period	12/18/11 → 12/21/11

Keywords

Multivariate Bag-of-Patterns
Stacked Bags-of-Patterns
classification
inverse frequency
multivariate time series

ASJC Scopus subject areas

Computer Science Applications
Human-Computer Interaction

Access to Document

10.1109/ICMLA.2011.46

Cite this

Ordóñez, P., Armstrong, T., Oates, T., & Fackler, J. (2011). Classification of patients using novel multivariate time series representations of physiological data. In Proceedings - 10th International Conference on Machine Learning and Applications, ICMLA 2011 (pp. 172-179). Article 6147668 (Proceedings - 10th International Conference on Machine Learning and Applications, ICMLA 2011; Vol. 2). https://doi.org/10.1109/ICMLA.2011.46

Classification of patients using novel multivariate time series representations of physiological data. / Ordóñez, Patricia; Armstrong, Tom; Oates, Tim et al.
Proceedings - 10th International Conference on Machine Learning and Applications, ICMLA 2011. 2011. p. 172-179 6147668 (Proceedings - 10th International Conference on Machine Learning and Applications, ICMLA 2011; Vol. 2).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Ordóñez, P, Armstrong, T, Oates, T & Fackler, J 2011, Classification of patients using novel multivariate time series representations of physiological data. in Proceedings - 10th International Conference on Machine Learning and Applications, ICMLA 2011., 6147668, Proceedings - 10th International Conference on Machine Learning and Applications, ICMLA 2011, vol. 2, pp. 172-179, 10th International Conference on Machine Learning and Applications, ICMLA 2011, Honolulu, HI, United States, 12/18/11. https://doi.org/10.1109/ICMLA.2011.46

Ordóñez P, Armstrong T, Oates T, Fackler J. Classification of patients using novel multivariate time series representations of physiological data. In Proceedings - 10th International Conference on Machine Learning and Applications, ICMLA 2011. 2011. p. 172-179. 6147668. (Proceedings - 10th International Conference on Machine Learning and Applications, ICMLA 2011). doi: 10.1109/ICMLA.2011.46

@inproceedings{34d0fb9e41b24dc2a692d9f40be5f92a,

title = "Classification of patients using novel multivariate time series representations of physiological data",

abstract = "In this paper we present two novel multivariate time series representations to classify physiological data of different lengths. The representations may be applied to any group of multivariate time series data that examine the state or health of an entity. Multivariate Bag-of-Patterns and Stacked Bags of-Patterns improve on their univariate counterpart, inspired by the bag-of-words model, by using multiple time series and analyzing the data in a multivariate fashion. We also borrow techniques from the natural language processing domain such as term frequency and inverse document frequency to improve classification accuracy. We introduce a technique named inverse frequency and present experimental results on classifying patients who have experienced acute episodes of hypotension.",

keywords = "Multivariate Bag-of-Patterns, Stacked Bags-of-Patterns, classification, inverse frequency, multivariate time series",

author = "Patricia Ord{\'o}{\~n}ez and Tom Armstrong and Tim Oates and Jim Fackler",

year = "2011",

doi = "10.1109/ICMLA.2011.46",

language = "English (US)",

isbn = "9780769546070",

series = "Proceedings - 10th International Conference on Machine Learning and Applications, ICMLA 2011",

pages = "172--179",

booktitle = "Proceedings - 10th International Conference on Machine Learning and Applications, ICMLA 2011",

note = "10th International Conference on Machine Learning and Applications, ICMLA 2011 ; Conference date: 18-12-2011 Through 21-12-2011",

}

TY - GEN

T1 - Classification of patients using novel multivariate time series representations of physiological data

AU - Ordóñez, Patricia

AU - Armstrong, Tom

AU - Oates, Tim

AU - Fackler, Jim

PY - 2011

Y1 - 2011

N2 - In this paper we present two novel multivariate time series representations to classify physiological data of different lengths. The representations may be applied to any group of multivariate time series data that examine the state or health of an entity. Multivariate Bag-of-Patterns and Stacked Bags of-Patterns improve on their univariate counterpart, inspired by the bag-of-words model, by using multiple time series and analyzing the data in a multivariate fashion. We also borrow techniques from the natural language processing domain such as term frequency and inverse document frequency to improve classification accuracy. We introduce a technique named inverse frequency and present experimental results on classifying patients who have experienced acute episodes of hypotension.

AB - In this paper we present two novel multivariate time series representations to classify physiological data of different lengths. The representations may be applied to any group of multivariate time series data that examine the state or health of an entity. Multivariate Bag-of-Patterns and Stacked Bags of-Patterns improve on their univariate counterpart, inspired by the bag-of-words model, by using multiple time series and analyzing the data in a multivariate fashion. We also borrow techniques from the natural language processing domain such as term frequency and inverse document frequency to improve classification accuracy. We introduce a technique named inverse frequency and present experimental results on classifying patients who have experienced acute episodes of hypotension.

KW - Multivariate Bag-of-Patterns

KW - Stacked Bags-of-Patterns

KW - classification

KW - inverse frequency

KW - multivariate time series

UR - http://www.scopus.com/inward/record.url?scp=84857879278&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84857879278&partnerID=8YFLogxK

U2 - 10.1109/ICMLA.2011.46

DO - 10.1109/ICMLA.2011.46

M3 - Conference contribution

AN - SCOPUS:84857879278

SN - 9780769546070

T3 - Proceedings - 10th International Conference on Machine Learning and Applications, ICMLA 2011

SP - 172

EP - 179

BT - Proceedings - 10th International Conference on Machine Learning and Applications, ICMLA 2011

T2 - 10th International Conference on Machine Learning and Applications, ICMLA 2011

Y2 - 18 December 2011 through 21 December 2011

ER -

Classification of patients using novel multivariate time series representations of physiological data

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this