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

Interpretable graph-based models on multimodal biomedical data integration: a technical review and benchmarking.

Nature communications·2026
Same author

Deep learning for apple leaf disease diagnosis: a comparative study with convolutional neural networks and transformers.

Scientific reports·2026
Same author

ECG-Adapt: A Novel Framework for Robust Electrocardiogram Classification Across Diverse Populations and Recording Conditions.

IEEE transactions on bio-medical engineering·2026
Same author

An explainable transformer model for Alzheimer's disease detection using retinal imaging.

Scientific reports·2025
Same author

Factors influencing the adoption of Internet of Medical Things for remote patient monitoring: A systematic literature review.

Computers in biology and medicine·2025
Same author

Post-Cardiac arrest outcome prediction using machine learning: A systematic review and meta-analysis.

International journal of medical informatics·2024

Related Experiment Video

Updated: Jan 8, 2026

Author Spotlight: Advancing the Study of Brain-Heart Interplay with a Comprehensive EEGLAB Plugin for Multimodal Signal Analysis
08:22

Author Spotlight: Advancing the Study of Brain-Heart Interplay with a Comprehensive EEGLAB Plugin for Multimodal Signal Analysis

Published on: April 26, 2024

2.9K

Spectral anomaly detection in physiological time-series data: A systematic review.

Emil Mittag1, Farshid Hajati1, Raymond Chiong2

  • 1School of Science and Technology, University of New England, Armidale, NSW, 2350, Australia.

International Journal of Medical Informatics
|December 18, 2025
PubMed
Summary
This summary is machine-generated.

Unsupervised transformer models excel at detecting anomalies in spectral data like ECG and EEG. These advanced machine learning methods achieve high accuracy without needing labeled data, improving diagnostic speed and patient outcomes.

Keywords:
Anomaly detectionCardiac arrestDeep learningECGEEGEpilepsyMachine learningSpectral data

More Related Videos

Software for Analysis of Heart Rate and Blood Pressure Time-series Data from the Valsalva Maneuver
14:28

Software for Analysis of Heart Rate and Blood Pressure Time-series Data from the Valsalva Maneuver

Published on: June 27, 2025

926
Computer-based Multitaper Spectrogram Program for Electroencephalographic Data
04:13

Computer-based Multitaper Spectrogram Program for Electroencephalographic Data

Published on: November 13, 2019

12.8K

Related Experiment Videos

Last Updated: Jan 8, 2026

Author Spotlight: Advancing the Study of Brain-Heart Interplay with a Comprehensive EEGLAB Plugin for Multimodal Signal Analysis
08:22

Author Spotlight: Advancing the Study of Brain-Heart Interplay with a Comprehensive EEGLAB Plugin for Multimodal Signal Analysis

Published on: April 26, 2024

2.9K
Software for Analysis of Heart Rate and Blood Pressure Time-series Data from the Valsalva Maneuver
14:28

Software for Analysis of Heart Rate and Blood Pressure Time-series Data from the Valsalva Maneuver

Published on: June 27, 2025

926
Computer-based Multitaper Spectrogram Program for Electroencephalographic Data
04:13

Computer-based Multitaper Spectrogram Program for Electroencephalographic Data

Published on: November 13, 2019

12.8K

Area of Science:

  • Health Informatics
  • Machine Learning
  • Signal Processing

Background:

  • Anomaly detection in Electrocardiogram (ECG) and Electroencephalogram (EEG) spectra is crucial for diagnosing critical conditions like cardiac arrest and epileptic seizures.
  • Machine learning, particularly deep learning, offers a pathway to automate and accelerate anomaly detection in physiological time-series data.
  • Timely diagnosis through automated anomaly detection can significantly improve patient treatment outcomes.

Purpose of the Study:

  • To systematically review machine learning applications for automated anomaly detection in ECG and EEG spectral data.
  • To identify and compare the effectiveness of various machine learning methods in spectral anomaly detection.
  • To provide recommendations on optimal methods for spectral anomaly detection, potentially applicable across different domains.

Main Methods:

  • A systematic literature review was conducted following PRISMA guidelines, searching major scientific databases (Web of Science, Scopus, PubMed, IEEE Xplore) up to October 2025.
  • Studies focusing on machine learning, including deep learning, applied to ECG and EEG spectra for anomaly detection were selected.
  • Articles reporting an Area Under the Curve (AUC), accuracy, or F1 score exceeding 0.95 were included in the final analysis.

Main Results:

  • The review included 65 articles from an initial pool of 519 search results.
  • Unsupervised machine learning methods, including variational autoencoders, generative adversarial networks, diffusion models, and transformers, achieved performance rates of 97%-99%.
  • Traditional models like isolation forest and support vector data description showed lower performance, plateauing between 90%-95%.

Conclusions:

  • Unsupervised transformer models demonstrated superior performance for anomaly detection in spectral data.
  • These models achieve high accuracy without the need for labeled datasets, making them highly efficient.
  • The findings suggest that unsupervised transformers may be the most suitable method for spectral anomaly detection and could be transferable to other data domains.