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

You might also read

Related Articles

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

Sort by
Same author

Consensus statement from the 2025 Delphi panel on cerebral microdialysis in critical care.

Critical care (London, England)·2026
Same author

Innovation in intensive care: a framework to turn ideas and concepts into actionable solutions.

Intensive care medicine·2026
Same author

Active Rehabilitation Technologies for Post-Stroke Patients.

Biosensors·2026
Same author

Wearable intelligent throat enables natural speech in stroke patients with dysarthria.

Nature communications·2026
Same author

Beyond overconfidence: Embedding curiosity and humility for ethical medical AI.

PLOS digital health·2026
Same author

Mecp2 deficiency impairs microscale cortical network topology and dynamics in a Rett syndrome mouse model.

bioRxiv : the preprint server for biology·2025

Related Experiment Video

Updated: Nov 9, 2025

Cortical Source Analysis of High-Density EEG Recordings in Children
09:32

Cortical Source Analysis of High-Density EEG Recordings in Children

Published on: June 30, 2014

21.6K

DeepClean: Self-Supervised Artefact Rejection for Intensive Care Waveform Data Using Deep Generative Learning.

Tom Edinburgh1, Peter Smielewski2, Marek Czosnyka2

  • 1Department of Applied Mathematics and Theoretical Physics, University of Cambridge, Cambridge, UK. te269@cam.ac.uk.

Acta Neurochirurgica. Supplement
|April 11, 2021
PubMed
Summary
This summary is machine-generated.

DeepClean, a self-supervised deep learning system, effectively detects and removes artefacts from intensive care unit (ICU) physiological data. This AI approach improves data accuracy for clinical assessments and reduces false alarms.

Keywords:
Artefact detectionArterial blood pressureDeep generative modelsPhysiological waveformsVariational autoencoder

More Related Videos

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding
11:25

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Published on: July 26, 2013

43.7K
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.4K

Related Experiment Videos

Last Updated: Nov 9, 2025

Cortical Source Analysis of High-Density EEG Recordings in Children
09:32

Cortical Source Analysis of High-Density EEG Recordings in Children

Published on: June 30, 2014

21.6K
Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding
11:25

Simultaneous Scalp Electroencephalography EEG, Electromyography EMG, and Whole-body Segmental Inertial Recording for Multi-modal Neural Decoding

Published on: July 26, 2013

43.7K
Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.4K

Area of Science:

  • Biomedical Engineering
  • Artificial Intelligence in Medicine
  • Critical Care Medicine

Background:

  • Waveform physiological data are crucial for treating critically ill patients.
  • Data artefacts in intensive care unit (ICU) recordings can lead to clinical assessment bias and increased false positive rates for alarms.
  • Accurate artefact removal is essential for reliable clinical decision-making and research.

Purpose of the Study:

  • To develop and evaluate DeepClean, a self-supervised system for detecting artefacts in waveform physiological data.
  • To assess the performance of DeepClean using invasive arterial blood pressure data.
  • To compare DeepClean's efficacy against traditional artefact detection methods.

Main Methods:

  • Developed DeepClean, a prototype self-supervised artefact detection system utilizing a convolutional variational autoencoder deep neural network.
  • Trained the system using only 'good' physiological data, avoiding manual annotation.
  • Tested the system on invasive arterial blood pressure recordings, evaluating its ability to detect artefacts within 10s samples.

Main Results:

  • DeepClean achieved approximately 90% sensitivity and specificity in detecting artefacts in 10s data samples.
  • The system accurately identified specific regions of artefacts within the data.
  • DeepClean significantly outperformed a principal component analysis approach in both signal reconstruction and artefact detection.

Conclusions:

  • DeepClean offers a robust, self-supervised method for artefact detection in ICU waveform data.
  • The system enhances data reliability for clinical assessment and alarm systems.
  • DeepClean's generative model capabilities also allow for potential imputation of missing data.