Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Pulse rhythm01:30

Pulse rhythm

985
Pulse rhythm refers to the pattern of pulsations within specific intervals, offering valuable insights into the regularity or irregularity of the heart's beats as observed through the pattern of pulsation within specific intervals. A regular pulse exhibits a consistent heart rate with uniform waveforms and pulsation force, variations of which can be classified as normal, weak, or bounding.
Conversely, an irregular pulse pattern is termed dysrhythmia, stemming from disruptions in cardiac...
985

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Reduced Plasma Aβ Peptides but Stable NfL and GFAP in Major Depressive Disorder.

International journal of molecular sciences·2026
Same author

Explainable machine learning with bayesian hyper-optimization for predicting cognitive impairment from longitudinal nursing home data.

Scientific reports·2026
Same author

Artificial intelligence-based identification of key risk factors for long COVID from early clinical data.

Journal of infection and public health·2025
Same author

Leaflet thrombosis automatic identification in transcatheter aortic valves using 4DCT.

Computers in biology and medicine·2025
Same author

Automated transcatheter heart valve 4DCT-based deformation assessment throughout the cardiac cycle: Towards enhanced long-term durability.

International journal of medical informatics·2025
Same author

Coronary Artery Disease in Very Young Women: Risk Factors and Prognostic Insights from Extended Follow-Up.

Journal of cardiovascular development and disease·2025

Related Experiment Video

Updated: Oct 13, 2025

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
07:31

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack

Published on: May 15, 2020

7.3K

Using Explainable Machine Learning to Improve Intensive Care Unit Alarm Systems.

José A González-Nóvoa1, Laura Busto1, Juan J Rodríguez-Andina2

  • 1Cardiovascular Research Group, Galicia Sur Health Research Institute (IIS Galicia Sur), 36213 Vigo, Spain.

Sensors (Basel, Switzerland)
|November 13, 2021
PubMed
Summary
This summary is machine-generated.

This study introduces an explainable machine learning model to predict Intensive Care Unit (ICU) patient mortality. The model optimizes ICU alarms by identifying key risk factors across different age groups, improving patient monitoring.

Keywords:
Intensive Care UnitMIMICalarmsexplainable machine learningmachine learningpatient monitoringsensors

More Related Videos

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

7.0K

Related Experiment Videos

Last Updated: Oct 13, 2025

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack
07:31

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack

Published on: May 15, 2020

7.3K
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

7.0K

Area of Science:

  • Medical Informatics
  • Artificial Intelligence in Healthcare
  • Clinical Data Analysis

Background:

  • Intensive Care Units (ICUs) generate vast amounts of patient data, challenging manual analysis, especially during high-demand periods like the COVID-19 pandemic.
  • Effective patient monitoring and timely alerts are crucial for critical care, but current alarm systems may not optimally utilize available data.
  • Automatic data analysis offers significant potential for enhancing patient monitoring and optimizing clinical decision-making in ICUs.

Purpose of the Study:

  • To develop and evaluate an explainable machine learning methodology for predicting patient mortality in the Intensive Care Unit (ICU).
  • To optimize ICU alarm systems by identifying critical clinical features and their impact thresholds on patient outcomes.
  • To provide insights into age-specific mortality predictors using explainable AI techniques.

Main Methods:

  • A methodology employing age-stratification and boosting classifiers was developed for mortality prediction.
  • Shapley Additive Explanations (SHAP) was utilized to interpret model predictions and identify key mortality-influencing features.
  • The MIMIC-III database, a comprehensive ICU patient research dataset, was used for model evaluation.

Main Results:

  • The proposed model achieved high accuracy in predicting ICU mortality across different age groups, with Area Under the Receiver Operating Characteristic Curve (AUROC) values ranging from 0.883 to 0.961.
  • SHAP analysis successfully identified the most impactful clinical features for mortality prediction within specific age strata (18-45, 45-65, 65-85, 85+).
  • The study identified specific thresholds for clinical features indicating a negative impact on patient health, crucial for alarm system refinement.

Conclusions:

  • Explainable machine learning, specifically using SHAP, can effectively predict ICU mortality and provide actionable insights for clinical practice.
  • The developed methodology enables the optimization of ICU alarm systems by highlighting critical variables and warning thresholds.
  • This approach supports healthcare personnel by offering data-driven insights for improved patient monitoring and timely intervention.