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

Unmixing the Neck: Accurate Jugular Venous Pulse Detection From Wearable PPG.

IEEE journal of biomedical and health informatics·2026
Same author

Evaluating the Pulse Rate Estimation Performance of the DS-EWMA Algorithm.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
Same author

A TinyML Motion-Based Embedded Cough Detection System.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
Same author

Heart Rate Estimation from Neck Photoplethysmography using FFT-Based Scoring and a Shallow Neural Network.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2025
Same author

Clinical Validation of Respiratory Rate Estimation Using Acoustic Signals from a Wearable Device.

Journal of clinical medicine·2024
Same author

A novel computational signal processing framework towards multimodal vital signs extraction using neck-worn wearable devices.

Scientific reports·2024

Related Experiment Video

Updated: Jun 28, 2025

Evaluation of Commercial-Off-The-Shelf Wrist Wearables to Estimate Stress on Students
12:51

Evaluation of Commercial-Off-The-Shelf Wrist Wearables to Estimate Stress on Students

Published on: June 16, 2018

7.5K

A comparative study in class imbalance mitigation when working with physiological signals.

Rawan S Abdulsadig1, Esther Rodriguez-Villegas1

  • 1Wearable Technologies Lab, Department of Electrical and Electronic Engineering, Imperial College London, London, United Kingdom.

Frontiers in Digital Health
|April 10, 2024
PubMed
Summary

Class imbalance in medical event detection, like apnoea, can be addressed with rebalancing algorithms. Random undersampling improved sensitivity for detecting apnoea from PPG signals but may reduce overall accuracy.

Keywords:
apneaclass imbalancemachine learningphysiological signalssudden unexpected death in epilepsy (SUDEP)

More Related Videos

A Single-Channel and Non-Invasive Wearable Brain-Computer Interface for Industry and Healthcare
06:34

A Single-Channel and Non-Invasive Wearable Brain-Computer Interface for Industry and Healthcare

Published on: July 7, 2023

2.3K
Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface
11:54

Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface

Published on: May 8, 2021

4.4K

Related Experiment Videos

Last Updated: Jun 28, 2025

Evaluation of Commercial-Off-The-Shelf Wrist Wearables to Estimate Stress on Students
12:51

Evaluation of Commercial-Off-The-Shelf Wrist Wearables to Estimate Stress on Students

Published on: June 16, 2018

7.5K
A Single-Channel and Non-Invasive Wearable Brain-Computer Interface for Industry and Healthcare
06:34

A Single-Channel and Non-Invasive Wearable Brain-Computer Interface for Industry and Healthcare

Published on: July 7, 2023

2.3K
Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface
11:54

Real-Time Proxy-Control of Re-Parameterized Peripheral Signals using a Close-Loop Interface

Published on: May 8, 2021

4.4K

Area of Science:

  • Biomedical Engineering
  • Machine Learning in Healthcare
  • Signal Processing

Background:

  • Class imbalance presents a significant challenge in detecting rare medical events such as apnoea.
  • Photoplethysmography (PPG) signals offer a potential data source for non-invasive apnoea detection.
  • Class rebalancing techniques are crucial for mitigating imbalance issues in medical classification tasks.

Purpose of the Study:

  • To investigate the effectiveness of 10 data-level class imbalance mitigation methods for detecting apnoea events.
  • To build and evaluate a Random Forest (RF) model using imbalanced PPG data.
  • To assess the impact of feature-space transformations (PCA, KernelPCA) on class rebalancing performance.

Main Methods:

  • Evaluated ten class imbalance mitigation techniques: RandUS, RandOS, CNNUS, ENNUS, TomekUS, SMOTE, BLSMOTE, ADASYN, SMOTETomek, and SMOTEENN.
  • Employed Random Forest (RF) as the classification model.
  • Explored Principal Component Analysis (PCA) and KernelPCA for feature-space transformation.

Main Results:

  • Random Undersampling (RandUS) demonstrated the most significant improvement in sensitivity, increasing it by up to 11%.
  • While RandUS enhanced sensitivity, it potentially decreased overall accuracy due to data reduction.
  • Data augmentation techniques, particularly with subject dependencies, require further research and development.

Conclusions:

  • RandUS is a viable strategy for enhancing apnoea detection sensitivity from PPG signals.
  • Careful consideration of the trade-off between sensitivity and accuracy is necessary when applying undersampling methods.
  • Advanced data augmentation methods are needed to effectively handle imbalanced medical datasets with inherent subject variability.