research-article
Authors: Ping Chang, Huayu Li, Stuart F. Quan, Shuyang Lu, + 3, Shu-Fen Wung, Janet Roveda, and Ao Li (Less)
Volume 246, Issue C
Published: 25 June 2024 Publication History
- 0citation
- 0
- Downloads
Metrics
Total Citations0Total Downloads0Last 12 Months0
Last 6 weeks0
New Citation Alert added!
This alert has been successfully added and will be sent to:
You will be notified whenever a record that you have chosen has been cited.
To manage your alert preferences, click on the button below.
Manage my Alerts
New Citation Alert!
Please log in to your account
- View Options
- References
- Media
- Tables
- Share
Abstract
Background and Objective
Vital sign monitoring in the Intensive Care Unit (ICU) is crucial for enabling prompt interventions for patients. This underscores the need for an accurate predictive system. Therefore, this study proposes a novel deep learning approach for forecasting Heart Rate (HR), Systolic Blood Pressure (SBP), and Diastolic Blood Pressure (DBP) in the ICU.
Methods
We extracted 24, 886 ICU stays from the MIMIC-III database which contains data from over 46 thousand patients, to train and test the model. The model proposed in this study, Transformer-based Diffusion Probabilistic Model for Sparse Time Series Forecasting (TDSTF), merges Transformer and diffusion models to forecast vital signs. The TDSTF model showed state-of-the-art performance in predicting vital signs in the ICU, outperforming other models' ability to predict distributions of vital signs and being more computationally efficient. The code is available at https://github.com/PingChang818/TDSTF.
Results
The results of the study showed that TDSTF achieved a Standardized Average Continuous Ranked Probability Score (SACRPS) of 0.4438 and a Mean Squared Error (MSE) of 0.4168, an improvement of 18.9% and 34.3% over the best baseline model, respectively. The inference speed of TDSTF is more than 17 times faster than the best baseline model.
Conclusion
TDSTF is an effective and efficient solution for forecasting vital signs in the ICU, and it shows a significant improvement compared to other models in the field.
Highlights
•
The TDSTF model forecasts vital signs based on all events in the MIMIC-III dataset.
•
The proposed model captures temporal patterns in both slow and sudden changes.
•
The MSE of the model improves by 34.3% over the best baseline model.
•
The SACRPS of the model improves by 18.9% over the best baseline model.
•
The inference speed of the model is more than 17 times faster than the best baseline model.
References
[1]
Tsuneaki Kenzaka, Masanobu Okayama, Shigehiro Kuroki, Miho f*ckui, Shinsuke Yahata, Hiroki Hayashi, Akihito Kitao, Daisuke Sugiyama, Eiji Kajii, Masayoshi Hashimoto, Importance of vital signs to the early diagnosis and severity of sepsis: association between vital signs and sequential organ failure assessment score in patients with sepsis, Intern. Med. 51 (8) (2012) 871–876.
[2]
Joo Heung Yoon, Lidan Mu, Lujie Chen, Artur Dubrawski, Marilyn Hravnak, Michael R. Pinsky, Gilles Clermont, Predicting tachycardia as a surrogate for instability in the intensive care unit, J. Clin. Monit. Comput. 33 (2019) 973–985.
[3]
Christian P. Subbe, Michael Kruger, Peter Rutherford, L. Gemmel, Validation of a modified early warning score in medical admissions, Q. J. Med. 94 (10) (2001) 521–526.
[4]
Daniel I. Sessler, Bernd Saugel, Beyond ‘failure to rescue’: the time has come for continuous ward monitoring, Br. J. Anaesth. 122 (3) (2019) 304–306.
[5]
Alexa K. Doig, Frank A. Drews, Maureen R. Keefe, Informing the design of hemodynamic monitoring displays, CIN, Comput. Inform. Nurs. 29 (12) (2011) 706–713.
[6]
Sarah A. Collins, Lena Mamykina, Desmond Jordan, Dan M. Stein, Alisabeth Shine, Paul Reyfman, David Kaufman, In search of common ground in handoff documentation in an intensive care unit, J. Biomed. Inform. 45 (2) (2012) 307–315.
[7]
Ævar Örn Kristinsson, Ying Gu, Søren M. Rasmussen, Jesper Mølgaard, Camilla Haahr-Raunkjær, Christian S. Meyhoff, Eske K. Aasvang, Helge B.D. Sørensen, Prediction of serious outcomes based on continuous vital sign monitoring of high-risk patients, Comput. Biol. Med. 147 (2022).
[8]
Marzyeh Ghassemi, Marco Pimentel, Tristan Naumann, Thomas Brennan, David Clifton, Peter Szolovits, Mengling Feng, A multivariate timeseries modeling approach to severity of illness assessment and forecasting in ICU with sparse, heterogeneous clinical data, in: Proceedings of the AAAI Conference on Artificial Intelligence, vol. 29, 2015.
[9]
Sindhu Tipirneni, Chandan K. Reddy, Self-supervised transformer for sparse and irregularly sampled multivariate clinical time-series, ACM Trans. Knowl. Discov. Data (TKDD) 16 (6) (2022) 1–17.
[10]
Inigo Jauregi Unanue, Ehsan Zare Borzeshi, Massimo Piccardi, Recurrent neural networks with specialized word embeddings for health-domain named-entity recognition, J. Biomed. Inform. 76 (2017) 102–109.
Digital Library
[11]
H. Ij, Statistics versus machine learning, Nat. Methods 15 (4) (2018) 233.
[12]
Shiyu Liu, Jia Yao, Mehul Motani, Early prediction of vital signs using generative boosting via LSTM networks, in: 2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), IEEE, 2019, pp. 437–444.
[13]
Shamsul Masum, John P. Chiverton, Ying Liu, Branislav Vuksanovic, Investigation of machine learning techniques in forecasting of blood pressure time series data, in: Artificial Intelligence XXXVI: 39th SGAI International Conference on Artificial Intelligence, Proceedings, vol. 39, AI 2019, Cambridge, UK, December 17–19, 2019, Springer, 2019, pp. 269–282.
[14]
Xiaoli Liu, Tongbo Liu, Zhengbo Zhang, Po-Chih Kuo, Haoran Xu, Zhicheng Yang, Ke Lan, Peiyao Li, Zhenchao Ouyang, Yeuk Lam Ng, et al., Top-net prediction model using bidirectional long short-term memory and medical-grade wearable multisensor system for tachycardia onset: algorithm development study, JMIR Med. Inform. 9 (4) (2021).
[15]
Ratchakit Phetrittikun, Kerdkiat Suvirat, Thanakron Na Pattalung, Chanon Kongkamol, Thammasin Ingviya, Sitthichok Chaichulee, Temporal fusion transformer for forecasting vital sign trajectories in intensive care patients, in: 2021 13th Biomedical Engineering International Conference (BMEiCON), IEEE, 2021, pp. 1–5.
[16]
Kashif Rasul, Calvin Seward, Ingmar Schuster, Roland Vollgraf, Autoregressive denoising diffusion models for multivariate probabilistic time series forecasting, in: International Conference on Machine Learning, PMLR, 2021, pp. 8857–8868.
[17]
Yusuke Tashiro, Jiaming Song, Yang Song, Stefano Ermon, Csdi: conditional score-based diffusion models for probabilistic time series imputation, Adv. Neural Inf. Process. Syst. 34 (2021) 24804–24816.
[18]
Jascha Sohl-Dickstein, Eric Weiss, Niru Maheswaranathan, Surya Ganguli, Deep unsupervised learning using nonequilibrium thermodynamics, in: International Conference on Machine Learning, PMLR, 2015, pp. 2256–2265.
[19]
Wen, Ruofeng; Torkkola, Kari; Narayanaswamy, Balakrishnan; Madeka, Dhruv (2017): A multi-horizon quantile recurrent forecaster. arXiv preprint arXiv:1711.11053.
[20]
David Salinas, Valentin Flunkert, Jan Gasthaus, Tim Januschowski Deepar, Probabilistic forecasting with autoregressive recurrent networks, Int. J. Forecast. 36 (3) (2020) 1181–1191.
[21]
Yuyang Wang, Alex Smola, Danielle Maddix, Jan Gasthaus, Dean Foster, Tim Januschowski, Deep factors for forecasting, in: International Conference on Machine Learning, PMLR, 2019, pp. 6607–6617.
[22]
Vilde Jensen, Filippo Maria Bianchi, Stian Normann Anfinsen, Ensemble conformalized quantile regression for probabilistic time series forecasting, IEEE Trans. Neural Netw. Learn. Syst. (2022).
[23]
Ling Yang, Zhilong Zhang, Yang Song, Shenda Hong, Runsheng Xu, Yue Zhao, Wentao Zhang, Bin Cui, Ming-Hsuan Yang, Diffusion models: a comprehensive survey of methods and applications, ACM Comput. Surv. 56 (4) (2023) 1–39.
[24]
Jonathan Ho, Ajay Jain, Pieter Abbeel, Denoising diffusion probabilistic models, Adv. Neural Inf. Process. Syst. 33 (2020) 6840–6851.
[25]
Song, Jiaming; Meng, Chenlin; Ermon, Stefano (2020): Denoising diffusion implicit models. arXiv preprint arXiv:2010.02502.
[26]
Jacob Austin, Daniel D. Johnson, Jonathan Ho, Daniel Tarlow, Rianne van den Berg, Structured denoising diffusion models in discrete state-spaces, Adv. Neural Inf. Process. Syst. 34 (2021) 17981–17993.
[27]
Anand, Namrata; Achim, Tudor (2022): Protein structure and sequence generation with equivariant denoising diffusion probabilistic models. arXiv preprint arXiv:2205.15019.
[28]
Blau, Tsachi; Ganz, Roy; Kawar, Bahjat; Bronstein, Alex; Elad, Michael (2022): Threat model-agnostic adversarial defense using diffusion models. arXiv preprint arXiv:2207.08089.
[29]
van den Oord, Aaron; Dieleman, Sander; Zen, Heiga; Simonyan, Karen; Vinyals, Oriol; Graves, Alex; Kalchbrenner, Nal; Senior, Andrew; Kavukcuoglu, Koray (2016): Wavenet: a generative model for raw audio. arXiv preprint arXiv:1609.03499.
[30]
Kong, Zhifeng; Ping, Wei; Huang, Jiaji; Zhao, Kexin; Catanzaro, Bryan (2020): Diffwave: a versatile diffusion model for audio synthesis. arXiv preprint arXiv:2009.09761.
[31]
Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Łukasz Kaiser, Illia Polosukhin, Attention is all you need, Adv. Neural Inf. Process. Syst. 30 (2017).
Digital Library
[32]
Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D. Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al., Language models are few-shot learners, Adv. Neural Inf. Process. Syst. 33 (2020) 1877–1901.
[33]
Cheng Zhang, Judith Bütepage, Hedvig Kjellström, Stephan Mandt, Advances in variational inference, IEEE Trans. Pattern Anal. Mach. Intell. 41 (8) (2018) 2008–2026.
[34]
William Feller, On the theory of stochastic processes, with particular reference to applications, in: Selected Papers I, Springer, 2015, pp. 769–798.
[35]
Luo, Calvin (2022): Understanding diffusion models: a unified perspective. arXiv preprint arXiv:2208.11970.
[36]
Devlin, Jacob; Chang, Ming-Wei; Lee, Kenton; Toutanova, Kristina (2018): Bert: pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.
[37]
Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun, Deep residual learning for image recognition, in: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016, pp. 770–778.
[38]
Alistair Johnson, Tom Pollard, Roger Mark, MIMIC-III clinical database (version 1.4), PhysioNet 10 (C2XW26) (2016) 2.
[39]
Craig Lockwood, Tiffany Conroy-Hiller, Tamara Page, Vital signs, JBI Evid. Synth. 2 (6) (2004) 1–38.
[40]
Kingma, Diederik P.; Ba, Jimmy (2014): Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980.
[41]
Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al., Pytorch: an imperative style, high-performance deep learning library, Adv. Neural Inf. Process. Syst. 32 (2019).
[42]
Wim De Mulder, Steven Bethard, Marie-Francine Moens, A survey on the application of recurrent neural networks to statistical language modeling, Comput. Speech Lang. 30 (1) (2015) 61–98.
[43]
Ming Ding, Chang Zhou, Hongxia Yang, Jie Tang Cogltx, Applying bert to long texts, Adv. Neural Inf. Process. Syst. 33 (2020) 12792–12804.
Recommendations
- Heart rate complexity helps mortality prediction in the intensive care unit: A pilot study using artificial intelligence
Abstract Background
In intensive care units (ICUs), accurate mortality prediction is crucial for effective patient management and resource allocation. The Simplified Acute Physiology Score II (SAPS-2), though commonly used, relies heavily on comprehensive ...
Graphical abstract
Display Omitted
Highlights
- Novel AI-based score predicts mortality in ICU patients using heart rate and blood pressure curves.
- Heart rate complexity, age, and diastolic blood pressure are key predictors of in-hospital death.
- The proposed algorithm requires ...
Read More
- Adaptive blood glucose control for intensive care applications
Highlights A two-step glucose control strategy for intensive care unit (ICU) applications is proposed. During the initial stay (3h) in the ICU, blood glucose is controlled through a standard intensive care protocol, and data on blood glucose ...
Read More
- Design of Software Services for Computer-Based Infection Control and Antibiotic Management in the Intensive Care Unit
ETELEMED '09: Proceedings of the 2009 International Conference on eHealth, Telemedicine, and Social Medicine
The increase of nosocomial infections and antibiotic resistance at the Intensive Care Unit (ICU) requires new approaches for infection control and antibiotic management. Although large hospitals have already introduced database systems for automated ...
Read More
Comments
Information & Contributors
Information
Published In
Computer Methods and Programs in Biomedicine Volume 246, Issue C
Apr 2024
221 pages
ISSN:0169-2607
Issue’s Table of Contents
Elsevier B.V.
Publisher
Elsevier North-Holland, Inc.
United States
Publication History
Published: 25 June 2024
Author Tags
- Deep learning
- Time series forecasting
- Sparse data
- Vital signs
- ICU
Qualifiers
- Research-article
Contributors
Other Metrics
View Article Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
Total Citations
Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Other Metrics
View Author Metrics
Citations
View Options
View options
Get Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in
Full Access
Get this Publication
Media
Figures
Other
Tables
