Investigation of machine learning techniques in forecasting of blood pressure time series data

Shamsul Masum; John Chiverton; Ying Liu; Branislav Vuksanovic

doi:10.1007/978-3-030-34885-4_21

Investigation of machine learning techniques in forecasting of blood pressure time series data

Shamsul Masum, John Chiverton, Ying Liu, Branislav Vuksanovic

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

493 Downloads (Pure)

Abstract

The aim of this paper is to investigate different machine learning based forecasting techniques for forecasting of blood pressure and heart rate. Forecasting of blood pressure could potentially help a clinician to take preventative steps even before dangerous medical situations occur. This paper examines forecasting blood pressure 30 minutes in advance. Univariate and multivariate forecast models are considered. Different forecast strategies are also considered. To compare different forecast strategies, LSTM and BI-LSTM machine learning algorithms were included. Then univariate and multivariate LSTM, BI-LSTM and CNN machine learning algorithms were compared using the two best forecasting strategies. Comparative analysis between forecasting strategies suggest that MIMO and DIRMO forecast strategies provide the best accuracy in forecasting physiological time series data. Results also appear to show that multivariate forecast models for blood pressure and heart rate are more reliable compared to blood pressure alone. Comparative analysis between MIMO and DIRMO forecasting strategies appear to show that DIRMO is more reliable for both univariate and multivariate cases. Results also appear to show that the forecast model that uses BI-LSTM with the DIRMO strategy is the best overall.

Original language	English
Title of host publication	SGAI 2019: Artificial Intelligence XXXVI. Thirty-ninth SGAI International Conference on Artificial Intelligence
Subtitle of host publication	International Conference on Innovative Techniques and Applications of Artificial Intelligence
Editors	Max Bramer, Miltos Petridis
Publisher	Springer
Chapter	21
Pages	269-282
Number of pages	14
Volume	11927
ISBN (Electronic)	978-3-030-34885-4
ISBN (Print)	978-3-030-34884-7
DOIs	https://doi.org/10.1007/978-3-030-34885-4_21
Publication status	Published - 19 Nov 2019
Event	AI-2019 Thirty-ninth SGAI International Conference on Artificial Intelligence - Cambridge, United Kingdom Duration: 17 Dec 2019 → 19 Dec 2019 Conference number: 39 http://www.bcs-sgai.org/ai2019/

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
ISSN (Print)	0302-9743

Name	Lecture Notes in Artificial Intelligence
Publisher	Springer
ISSN (Print)	0302-9743

Conference

Conference	AI-2019 Thirty-ninth SGAI International Conference on Artificial Intelligence
Abbreviated title	SGAI
Country/Territory	United Kingdom
City	Cambridge
Period	17/12/19 → 19/12/19
Internet address	http://www.bcs-sgai.org/ai2019/

Keywords

Time series forecasting
Univariate data
Multivariate data
Forecast strategies
LSTM
BI-LSTM
CNN
Blood pressure
Heart rate

Access to Document

10.1007/978-3-030-34885-4_21

BCS_AI
This is a post-peer-review, pre-copyedit version of an article published in SGAI 2019: Artificial Intelligence XXXVI. The final authenticated version is available online at: http://dx.doi.org/10.1007/978-3-030-34885-4_21.
Accepted author manuscript (Post-print), 473 KB

Time Series Forecasting of Physiological Signals
Chiverton, J., Masum, S., Liu, Y., Vuksanovic, B. & Parker, M.
17/05/16 → …
Project: Research

Cite this

Masum, S., Chiverton, J., Liu, Y., & Vuksanovic, B. (2019). Investigation of machine learning techniques in forecasting of blood pressure time series data. In M. Bramer, & M. Petridis (Eds.), SGAI 2019: Artificial Intelligence XXXVI. Thirty-ninth SGAI International Conference on Artificial Intelligence: International Conference on Innovative Techniques and Applications of Artificial Intelligence (Vol. 11927, pp. 269-282). (Lecture Notes in Computer Science), (Lecture Notes in Artificial Intelligence). Springer. https://doi.org/10.1007/978-3-030-34885-4_21

Masum, Shamsul ; Chiverton, John ; Liu, Ying et al. / Investigation of machine learning techniques in forecasting of blood pressure time series data. SGAI 2019: Artificial Intelligence XXXVI. Thirty-ninth SGAI International Conference on Artificial Intelligence: International Conference on Innovative Techniques and Applications of Artificial Intelligence. editor / Max Bramer ; Miltos Petridis. Vol. 11927 Springer, 2019. pp. 269-282 (Lecture Notes in Computer Science). (Lecture Notes in Artificial Intelligence).

@inproceedings{82a4426490d14bda9604e00a84994cac,

title = "Investigation of machine learning techniques in forecasting of blood pressure time series data",

abstract = "The aim of this paper is to investigate different machine learning based forecasting techniques for forecasting of blood pressure and heart rate. Forecasting of blood pressure could potentially help a clinician to take preventative steps even before dangerous medical situations occur. This paper examines forecasting blood pressure 30 minutes in advance. Univariate and multivariate forecast models are considered. Different forecast strategies are also considered. To compare different forecast strategies, LSTM and BI-LSTM machine learning algorithms were included. Then univariate and multivariate LSTM, BI-LSTM and CNN machine learning algorithms were compared using the two best forecasting strategies. Comparative analysis between forecasting strategies suggest that MIMO and DIRMO forecast strategies provide the best accuracy in forecasting physiological time series data. Results also appear to show that multivariate forecast models for blood pressure and heart rate are more reliable compared to blood pressure alone. Comparative analysis between MIMO and DIRMO forecasting strategies appear to show that DIRMO is more reliable for both univariate and multivariate cases. Results also appear to show that the forecast model that uses BI-LSTM with the DIRMO strategy is the best overall.",

keywords = "Time series forecasting, Univariate data, Multivariate data, Forecast strategies, LSTM, BI-LSTM, CNN, Blood pressure, Heart rate",

author = "Shamsul Masum and John Chiverton and Ying Liu and Branislav Vuksanovic",

year = "2019",

month = nov,

day = "19",

doi = "10.1007/978-3-030-34885-4_21",

language = "English",

isbn = "978-3-030-34884-7",

volume = "11927",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "269--282",

editor = "Max Bramer and Miltos Petridis",

booktitle = "SGAI 2019: Artificial Intelligence XXXVI. Thirty-ninth SGAI International Conference on Artificial Intelligence",

note = "AI-2019 Thirty-ninth SGAI International Conference on Artificial Intelligence, SGAI ; Conference date: 17-12-2019 Through 19-12-2019",

url = "http://www.bcs-sgai.org/ai2019/",

}

Masum, S , Chiverton, J, Liu, Y & Vuksanovic, B 2019, Investigation of machine learning techniques in forecasting of blood pressure time series data. in M Bramer & M Petridis (eds), SGAI 2019: Artificial Intelligence XXXVI. Thirty-ninth SGAI International Conference on Artificial Intelligence: International Conference on Innovative Techniques and Applications of Artificial Intelligence. vol. 11927, Lecture Notes in Computer Science, Lecture Notes in Artificial Intelligence, Springer, pp. 269-282, AI-2019 Thirty-ninth SGAI International Conference on Artificial Intelligence, Cambridge, United Kingdom, 17/12/19. https://doi.org/10.1007/978-3-030-34885-4_21

Investigation of machine learning techniques in forecasting of blood pressure time series data. / Masum, Shamsul ; Chiverton, John; Liu, Ying et al.
SGAI 2019: Artificial Intelligence XXXVI. Thirty-ninth SGAI International Conference on Artificial Intelligence: International Conference on Innovative Techniques and Applications of Artificial Intelligence. ed. / Max Bramer; Miltos Petridis. Vol. 11927 Springer, 2019. p. 269-282 (Lecture Notes in Computer Science), (Lecture Notes in Artificial Intelligence).

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

TY - GEN

T1 - Investigation of machine learning techniques in forecasting of blood pressure time series data

AU - Masum, Shamsul

AU - Chiverton, John

AU - Liu, Ying

AU - Vuksanovic, Branislav

N1 - Conference code: 39

PY - 2019/11/19

Y1 - 2019/11/19

N2 - The aim of this paper is to investigate different machine learning based forecasting techniques for forecasting of blood pressure and heart rate. Forecasting of blood pressure could potentially help a clinician to take preventative steps even before dangerous medical situations occur. This paper examines forecasting blood pressure 30 minutes in advance. Univariate and multivariate forecast models are considered. Different forecast strategies are also considered. To compare different forecast strategies, LSTM and BI-LSTM machine learning algorithms were included. Then univariate and multivariate LSTM, BI-LSTM and CNN machine learning algorithms were compared using the two best forecasting strategies. Comparative analysis between forecasting strategies suggest that MIMO and DIRMO forecast strategies provide the best accuracy in forecasting physiological time series data. Results also appear to show that multivariate forecast models for blood pressure and heart rate are more reliable compared to blood pressure alone. Comparative analysis between MIMO and DIRMO forecasting strategies appear to show that DIRMO is more reliable for both univariate and multivariate cases. Results also appear to show that the forecast model that uses BI-LSTM with the DIRMO strategy is the best overall.

AB - The aim of this paper is to investigate different machine learning based forecasting techniques for forecasting of blood pressure and heart rate. Forecasting of blood pressure could potentially help a clinician to take preventative steps even before dangerous medical situations occur. This paper examines forecasting blood pressure 30 minutes in advance. Univariate and multivariate forecast models are considered. Different forecast strategies are also considered. To compare different forecast strategies, LSTM and BI-LSTM machine learning algorithms were included. Then univariate and multivariate LSTM, BI-LSTM and CNN machine learning algorithms were compared using the two best forecasting strategies. Comparative analysis between forecasting strategies suggest that MIMO and DIRMO forecast strategies provide the best accuracy in forecasting physiological time series data. Results also appear to show that multivariate forecast models for blood pressure and heart rate are more reliable compared to blood pressure alone. Comparative analysis between MIMO and DIRMO forecasting strategies appear to show that DIRMO is more reliable for both univariate and multivariate cases. Results also appear to show that the forecast model that uses BI-LSTM with the DIRMO strategy is the best overall.

KW - Time series forecasting

KW - Univariate data

KW - Multivariate data

KW - Forecast strategies

KW - LSTM

KW - BI-LSTM

KW - CNN

KW - Blood pressure

KW - Heart rate

U2 - 10.1007/978-3-030-34885-4_21

DO - 10.1007/978-3-030-34885-4_21

M3 - Conference contribution

SN - 978-3-030-34884-7

VL - 11927

T3 - Lecture Notes in Computer Science

SP - 269

EP - 282

BT - SGAI 2019: Artificial Intelligence XXXVI. Thirty-ninth SGAI International Conference on Artificial Intelligence

A2 - Bramer, Max

A2 - Petridis, Miltos

PB - Springer

T2 - AI-2019 Thirty-ninth SGAI International Conference on Artificial Intelligence

Y2 - 17 December 2019 through 19 December 2019

ER -

Masum S , Chiverton J, Liu Y, Vuksanovic B. Investigation of machine learning techniques in forecasting of blood pressure time series data. In Bramer M, Petridis M, editors, SGAI 2019: Artificial Intelligence XXXVI. Thirty-ninth SGAI International Conference on Artificial Intelligence: International Conference on Innovative Techniques and Applications of Artificial Intelligence. Vol. 11927. Springer. 2019. p. 269-282. (Lecture Notes in Computer Science). (Lecture Notes in Artificial Intelligence). doi: 10.1007/978-3-030-34885-4_21

Investigation of machine learning techniques in forecasting of blood pressure time series data

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Projects

Time Series Forecasting of Physiological Signals

Cite this