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

MedIntelliCare: neurodynamic-inspired AI for medical decision support by integrating retrieval-augmented generation with multimodal cognitive processing.

Cognitive neurodynamics·2026
Same author

C2DGCN: cross-connected distributive learning-enabled graph convolutional network for human emotion recognition using electroencephalography signal.

Cognitive neurodynamics·2025
See all related articles

Related Experiment Video

Updated: Jan 8, 2026

Author Spotlight: Advancing Labor Management Through Electromyometrial Imaging for Understanding Uterine Contractions
08:07

Author Spotlight: Advancing Labor Management Through Electromyometrial Imaging for Understanding Uterine Contractions

Published on: May 26, 2023

1.7K

Multi-DNBiTM: preterm labor prediction from electrohysterography signals using multi-head attention-enabled deep

Puja Cholke1, Umar M Mulani2, Ashutosh Madhukar Kulkarni3

  • 1Department of Information Technology Vishwakarma Institute of Technology, Bibwewadi, Pune, Maharashtra, India.

Computer Methods in Biomechanics and Biomedical Engineering
|December 24, 2025
PubMed
Summary

Accurate prediction of preterm labor is vital for neonatal survival. A new deep learning model, multi-DNBiTM, using Electrohysterography (EHG) signals, significantly improves prediction accuracy for preterm labor contractions.

Keywords:
Deep learningelectrohysterographymulti-level trainingpreterm labor predictiontime–frequency analysis

More Related Videos

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.2K
Preterm EEG: A Multimodal Neurophysiological Protocol
19:32

Preterm EEG: A Multimodal Neurophysiological Protocol

Published on: February 18, 2012

28.9K

Related Experiment Videos

Last Updated: Jan 8, 2026

Author Spotlight: Advancing Labor Management Through Electromyometrial Imaging for Understanding Uterine Contractions
08:07

Author Spotlight: Advancing Labor Management Through Electromyometrial Imaging for Understanding Uterine Contractions

Published on: May 26, 2023

1.7K
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

5.2K
Preterm EEG: A Multimodal Neurophysiological Protocol
19:32

Preterm EEG: A Multimodal Neurophysiological Protocol

Published on: February 18, 2012

28.9K

Area of Science:

  • Biomedical Engineering
  • Artificial Intelligence in Healthcare
  • Signal Processing

Background:

  • Timely preterm labor prediction is critical for neonatal outcomes and maternal care.
  • Conventional methods for predicting preterm labor using Electrohysterography (EHG) signals suffer from limitations like low sensitivity and robustness.
  • EHG signals offer high sensitivity for analyzing uterine contractions, presenting a promising approach for preterm labor detection.

Purpose of the Study:

  • To develop and validate a novel deep learning framework, multi-DNBiTM, for accurate preterm labor prediction.
  • To overcome the limitations of existing preterm labor prediction methods.
  • To enhance the analysis of subtle variations in uterine contractions for improved prediction.

Main Methods:

  • Implementation of a multi-head attention-enabled distributed neural network with bidirectional long short-term memory (multi-DNBiTM).
  • Utilization of Root mean energy-entropy deep features (RMEn2D) for detailed analysis of frequency subbands and minor contractions.
  • Application of multi-level training to enhance prediction accuracy by analyzing signals at different granularities.

Main Results:

  • The multi-DNBiTM model achieved high performance metrics: 96.93% accuracy, 98.45% sensitivity, and 98.19% specificity.
  • Demonstrated superior outcomes compared to prevailing approaches in preterm labor prediction.
  • Effectively captured intrinsic patterns and fine details in EHG signals through multi-head attention and multi-level training.

Conclusions:

  • The proposed multi-DNBiTM framework offers a robust and sensitive solution for preterm labor prediction using EHG signals.
  • The RMEn2D features and multi-head attention mechanism contribute to the model's enhanced predictive capabilities.
  • This advanced deep learning approach holds significant potential for improving clinical management of preterm labor and neonatal care.