Investigating the accuracy of neural networks for blood pressure prediction in the ICU
This paper reports on research which investigates the viability of artificial neural networks, used in an ICU environment, for predicting both systolic and diastolic blood pressure up to 1 h ahead. In this environment, patients often receive pharmacological intervention to increase or decrease blood...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Informatics in Medicine Unlocked |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352914825000231 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850135960360058880 |
|---|---|
| author | Charles J. Gillan Bartosz Gorecki |
| author_facet | Charles J. Gillan Bartosz Gorecki |
| author_sort | Charles J. Gillan |
| collection | DOAJ |
| description | This paper reports on research which investigates the viability of artificial neural networks, used in an ICU environment, for predicting both systolic and diastolic blood pressure up to 1 h ahead. In this environment, patients often receive pharmacological intervention to increase or decrease blood pressure. The physiological state of an ICU patient is therefore quite different to a hyper or hypotensive patient outside hospital, suggesting that predicting blood pressure in this environment is more challenging The work investigates whether building neural network architectures with multivariate input data is capable of predicting blood pressures in this environment. Our work uses skin temperature and heart rate readings in addition to systolic and diastolic blood pressure. Two types of neural network are explored are explored in this paper: an encoder-decoder long short-term memory architecture and, separately, a convolutional neural network architecture. The top-performing configuration, when using a 70 %–30 % train-test split of data, is a convolutional neural network model. This predicted systolic and diastolic blood pressures for a patient with an error of approximately 3.4 %. These results are at the same level of accuracy as work on blood pressure prediction outside the ICU environment. Our work shows that neural networks are a viable tool for short term prediction of arterial blood pressures in an ICU context. |
| format | Article |
| id | doaj-art-e0d59d8f49134448a17ead564c240ba4 |
| institution | OA Journals |
| issn | 2352-9148 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Informatics in Medicine Unlocked |
| spelling | doaj-art-e0d59d8f49134448a17ead564c240ba42025-08-20T02:31:16ZengElsevierInformatics in Medicine Unlocked2352-91482025-01-015510163510.1016/j.imu.2025.101635Investigating the accuracy of neural networks for blood pressure prediction in the ICUCharles J. Gillan0Bartosz Gorecki1Corresponding author.; School of Electrical and Electronic Engineering and Computer Science, Queen's University Belfast, Computer Science Building, 16A Malone Road Belfast, Northern Ireland, BT9 5BN, UKSchool of Electrical and Electronic Engineering and Computer Science, Queen's University Belfast, Computer Science Building, 16A Malone Road Belfast, Northern Ireland, BT9 5BN, UKThis paper reports on research which investigates the viability of artificial neural networks, used in an ICU environment, for predicting both systolic and diastolic blood pressure up to 1 h ahead. In this environment, patients often receive pharmacological intervention to increase or decrease blood pressure. The physiological state of an ICU patient is therefore quite different to a hyper or hypotensive patient outside hospital, suggesting that predicting blood pressure in this environment is more challenging The work investigates whether building neural network architectures with multivariate input data is capable of predicting blood pressures in this environment. Our work uses skin temperature and heart rate readings in addition to systolic and diastolic blood pressure. Two types of neural network are explored are explored in this paper: an encoder-decoder long short-term memory architecture and, separately, a convolutional neural network architecture. The top-performing configuration, when using a 70 %–30 % train-test split of data, is a convolutional neural network model. This predicted systolic and diastolic blood pressures for a patient with an error of approximately 3.4 %. These results are at the same level of accuracy as work on blood pressure prediction outside the ICU environment. Our work shows that neural networks are a viable tool for short term prediction of arterial blood pressures in an ICU context.http://www.sciencedirect.com/science/article/pii/S2352914825000231 |
| spellingShingle | Charles J. Gillan Bartosz Gorecki Investigating the accuracy of neural networks for blood pressure prediction in the ICU Informatics in Medicine Unlocked |
| title | Investigating the accuracy of neural networks for blood pressure prediction in the ICU |
| title_full | Investigating the accuracy of neural networks for blood pressure prediction in the ICU |
| title_fullStr | Investigating the accuracy of neural networks for blood pressure prediction in the ICU |
| title_full_unstemmed | Investigating the accuracy of neural networks for blood pressure prediction in the ICU |
| title_short | Investigating the accuracy of neural networks for blood pressure prediction in the ICU |
| title_sort | investigating the accuracy of neural networks for blood pressure prediction in the icu |
| url | http://www.sciencedirect.com/science/article/pii/S2352914825000231 |
| work_keys_str_mv | AT charlesjgillan investigatingtheaccuracyofneuralnetworksforbloodpressurepredictionintheicu AT bartoszgorecki investigatingtheaccuracyofneuralnetworksforbloodpressurepredictionintheicu |