Literature DB >> 26992068

Using Supervised Machine Learning to Classify Real Alerts and Artifact in Online Multisignal Vital Sign Monitoring Data.

Lujie Chen1, Artur Dubrawski, Donghan Wang, Madalina Fiterau, Mathieu Guillame-Bert, Eliezer Bose, Ata M Kaynar, David J Wallace, Jane Guttendorf, Gilles Clermont, Michael R Pinsky, Marilyn Hravnak.   

Abstract

OBJECTIVE: The use of machine-learning algorithms to classify alerts as real or artifacts in online noninvasive vital sign data streams to reduce alarm fatigue and missed true instability.
DESIGN: Observational cohort study.
SETTING: Twenty-four-bed trauma step-down unit. PATIENTS: Two thousand one hundred fifty-three patients. INTERVENTION: Noninvasive vital sign monitoring data (heart rate, respiratory rate, peripheral oximetry) recorded on all admissions at 1/20 Hz, and noninvasive blood pressure less frequently, and partitioned data into training/validation (294 admissions; 22,980 monitoring hours) and test sets (2,057 admissions; 156,177 monitoring hours). Alerts were vital sign deviations beyond stability thresholds. A four-member expert committee annotated a subset of alerts (576 in training/validation set, 397 in test set) as real or artifact selected by active learning, upon which we trained machine-learning algorithms. The best model was evaluated on test set alerts to enact online alert classification over time.
MEASUREMENTS AND MAIN RESULTS: The Random Forest model discriminated between real and artifact as the alerts evolved online in the test set with area under the curve performance of 0.79 (95% CI, 0.67-0.93) for peripheral oximetry at the instant the vital sign first crossed threshold and increased to 0.87 (95% CI, 0.71-0.95) at 3 minutes into the alerting period. Blood pressure area under the curve started at 0.77 (95% CI, 0.64-0.95) and increased to 0.87 (95% CI, 0.71-0.98), whereas respiratory rate area under the curve started at 0.85 (95% CI, 0.77-0.95) and increased to 0.97 (95% CI, 0.94-1.00). Heart rate alerts were too few for model development.
CONCLUSIONS: Machine-learning models can discern clinically relevant peripheral oximetry, blood pressure, and respiratory rate alerts from artifacts in an online monitoring dataset (area under the curve > 0.87).

Entities:  

Mesh:

Year:  2016        PMID: 26992068      PMCID: PMC4911247          DOI: 10.1097/CCM.0000000000001660

Source DB:  PubMed          Journal:  Crit Care Med        ISSN: 0090-3493            Impact factor:   7.598


  27 in total

1.  A transform domain SVD filter for suppression of muscle noise artefacts in exercise ECG's.

Authors:  J S Paul; M R Reddy; V J Kumar
Journal:  IEEE Trans Biomed Eng       Date:  2000-05       Impact factor: 4.538

2.  Automated classification of patients with coronary artery disease using grayscale features from left ventricle echocardiographic images.

Authors:  U Rajendra Acharya; S Vinitha Sree; M Muthu Rama Krishnan; N Krishnananda; Shetty Ranjan; Pai Umesh; Jasjit S Suri
Journal:  Comput Methods Programs Biomed       Date:  2013-08-16       Impact factor: 5.428

3.  Machine Learning in Computer-aided Diagnosis of the Thorax and Colon in CT: A Survey.

Authors:  Kenji Suzuki
Journal:  IEICE Trans Inf Syst       Date:  2013-04-01

4.  Rx for ECG monitoring artifact.

Authors:  M Smith
Journal:  Crit Care Nurse       Date:  1984 Jan-Feb       Impact factor: 1.708

5.  Alarm fatigue: a patient safety concern.

Authors:  Sue Sendelbach; Marjorie Funk
Journal:  AACN Adv Crit Care       Date:  2013 Oct-Dec

6.  Data mining framework for fatty liver disease classification in ultrasound: a hybrid feature extraction paradigm.

Authors:  U Rajendra Acharya; S Vinitha Sree; Ricardo Ribeiro; Ganapathy Krishnamurthi; Rui Tato Marinho; Joao Sanches; Jasjit S Suri
Journal:  Med Phys       Date:  2012-07       Impact factor: 4.071

7.  Standardized database development for EEG epileptiform transient detection: EEGnet scoring system and machine learning analysis.

Authors:  Jonathan J Halford; Robert J Schalkoff; Jing Zhou; Selim R Benbadis; William O Tatum; Robert P Turner; Saurabh R Sinha; Nathan B Fountain; Amir Arain; Paul B Pritchard; Ekrem Kutluay; Gabriel Martz; Jonathan C Edwards; Chad Waters; Brian C Dean
Journal:  J Neurosci Methods       Date:  2012-11-19       Impact factor: 2.390

8.  Removing ECG noise from surface EMG signals using adaptive filtering.

Authors:  Guohua Lu; John-Stuart Brittain; Peter Holland; John Yianni; Alexander L Green; John F Stein; Tipu Z Aziz; Shouyan Wang
Journal:  Neurosci Lett       Date:  2009-06-25       Impact factor: 3.046

9.  Combining an expert-based medical entity recognizer to a machine-learning system: methods and a case study.

Authors:  Pierre Zweigenbaum; Thomas Lavergne; Natalia Grabar; Thierry Hamon; Sophie Rosset; Cyril Grouin
Journal:  Biomed Inform Insights       Date:  2013-08-01

10.  Insights into the problem of alarm fatigue with physiologic monitor devices: a comprehensive observational study of consecutive intensive care unit patients.

Authors:  Barbara J Drew; Patricia Harris; Jessica K Zègre-Hemsey; Tina Mammone; Daniel Schindler; Rebeca Salas-Boni; Yong Bai; Adelita Tinoco; Quan Ding; Xiao Hu
Journal:  PLoS One       Date:  2014-10-22       Impact factor: 3.240

View more
  18 in total

1.  Predictive Monitoring of Critical Cardiorespiratory Alarms in Neonates Under Intensive Care.

Authors:  Rohan Joshi; Zheng Peng; Xi Long; Loe Feijs; Peter Andriessen; Carola Van Pul
Journal:  IEEE J Transl Eng Health Med       Date:  2019-11-22       Impact factor: 3.316

2.  Intensive care medicine in 2050: NEWS for hemodynamic monitoring.

Authors:  Frederic Michard; Michael R Pinsky; Jean-Louis Vincent
Journal:  Intensive Care Med       Date:  2017-01-25       Impact factor: 17.440

3.  Dynamic and Personalized Risk Forecast in Step-Down Units. Implications for Monitoring Paradigms.

Authors:  Lujie Chen; Olufunmilayo Ogundele; Gilles Clermont; Marilyn Hravnak; Michael R Pinsky; Artur W Dubrawski
Journal:  Ann Am Thorac Soc       Date:  2017-03

4.  The rise of ward monitoring: opportunities and challenges for critical care specialists.

Authors:  Frederic Michard; Rinaldo Bellomo; Andreas Taenzer
Journal:  Intensive Care Med       Date:  2018-09-27       Impact factor: 17.440

5.  Automated anesthesia artifact analysis: can machines be trained to take out the garbage?

Authors:  Allan F Simpao; Olivia Nelson; Luis M Ahumada
Journal:  J Clin Monit Comput       Date:  2020-09-12       Impact factor: 2.502

Review 6.  A review of recent advances in data analytics for post-operative patient deterioration detection.

Authors:  Clemence Petit; Rick Bezemer; Louis Atallah
Journal:  J Clin Monit Comput       Date:  2017-08-21       Impact factor: 2.502

Review 7.  Applying machine learning to continuously monitored physiological data.

Authors:  Barret Rush; Leo Anthony Celi; David J Stone
Journal:  J Clin Monit Comput       Date:  2018-11-11       Impact factor: 2.502

Review 8.  Opportunities for machine learning to improve surgical ward safety.

Authors:  Tyler J Loftus; Patrick J Tighe; Amanda C Filiberto; Jeremy Balch; Gilbert R Upchurch; Parisa Rashidi; Azra Bihorac
Journal:  Am J Surg       Date:  2020-02-26       Impact factor: 2.565

Review 9.  A call to alarms: Current state and future directions in the battle against alarm fatigue.

Authors:  Marilyn Hravnak; Tiffany Pellathy; Lujie Chen; Artur Dubrawski; Anthony Wertz; Gilles Clermont; Michael R Pinsky
Journal:  J Electrocardiol       Date:  2018-07-29       Impact factor: 1.438

Review 10.  Predicting adverse hemodynamic events in critically ill patients.

Authors:  Joo H Yoon; Michael R Pinsky
Journal:  Curr Opin Crit Care       Date:  2018-06       Impact factor: 3.687

View more

北京卡尤迪生物科技股份有限公司 © 2022-2023.