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

Efficacy of oral sucrose for preterm infants undergoing point-of-care lung ultrasound: a randomized assessor-blinded controlled trial.

European journal of pediatrics·2026
Same author

Protocol for developing a national approach to surveillance and prevention for neonatal ventilator-associated pneumonia.

BMJ paediatrics open·2026
Same author

Association of NICU design on neurodevelopmental outcomes in preterm neonates born at less than 29 weeks' gestation: a retrospective observational study.

Journal of perinatology : official journal of the California Perinatal Association·2025
Same author

Lung ultrasound assessment of pulmonary effects of large patent ductus arteriosus in extremely preterm infants beyond the transitional period.

The ultrasound journal·2025
Same author

Association of lung ultrasound score with large patent ductus arteriosus in preterm neonates during the transitional period.

European journal of pediatrics·2025
Same author

Comparing Malposition and Complications Associated with Ultrasound-Guided versus Radiography-Guided Central Catheter Tip-Position in Neonates: A Systematic Review and Meta-Analysis.

Neonatology·2025
Same journal

The Need for Demonstrated Clinical Translational Evidence in Submissions to the IEEE Journal of Translational Engineering in Health and Medicine.

IEEE journal of translational engineering in health and medicine·2026
Same journal

Accuracy of Quantifying Hypotension During Surgery Using Physiological Sensor Data.

IEEE journal of translational engineering in health and medicine·2026
Same journal

Analyzing Gait Pattern Associated With Neuropsychiatric Symptoms in Parkinson's Disease by a Comprehensive Approach.

IEEE journal of translational engineering in health and medicine·2026
Same journal

Multimodal Patient-Specific Identification of Atrial Flutter Circuits From ECG Time Series Using Explainable Machine Learning.

IEEE journal of translational engineering in health and medicine·2026
Same journal

Innovative Wearable Platform for Synchronized Biosignals Acquisition: A Proof of Concept in a Cuff-Less Blood Pressure Monitoring Case Study.

IEEE journal of translational engineering in health and medicine·2026
Same journal

Development of a Realistic Physical Phantom for Laparoscopic and Robotic-Assisted Sacrocolpopexy Training and Associated.

IEEE journal of translational engineering in health and medicine·2026
See all related articles

Related Experiment Video

Updated: Jul 8, 2025

Protocol and Guidelines for Point-of-Care Lung Ultrasound in Diagnosing Neonatal Pulmonary Diseases Based on International Expert Consensus
06:15

Protocol and Guidelines for Point-of-Care Lung Ultrasound in Diagnosing Neonatal Pulmonary Diseases Based on International Expert Consensus

Published on: March 6, 2019

50.0K

An Interpretable Neonatal Lung Ultrasound Feature Extraction and Lung Sliding Detection System Using Object

Rodina Bassiouny1, Adel Mohamed2, Karthi Umapathy1

  • 1Department of Electrical, Computer, and Biomedical EngineeringToronto Metropolitan University Toronto ON M5B 2K3 Canada.

IEEE Journal of Translational Engineering in Health and Medicine
|December 13, 2023
PubMed
Summary
This summary is machine-generated.

This study developed an automated system using object detection to analyze neonatal lung ultrasound images, improving Pneumothorax (PTX) diagnosis. The AI system accurately identifies key lung features, aiding faster and more precise clinical decisions.

Keywords:
Hough transformLung ultrasoundM-modeRetinaNetautomatic lung sliding detectionfaster RCNNobject detection models

More Related Videos

International Expert Consensus and Recommendations for Neonatal Pneumothorax Ultrasound Diagnosis and Ultrasound-guided Thoracentesis Procedure
05:50

International Expert Consensus and Recommendations for Neonatal Pneumothorax Ultrasound Diagnosis and Ultrasound-guided Thoracentesis Procedure

Published on: March 12, 2020

13.7K
Point-of-Care Lung Ultrasound in Adults: Image Acquisition
09:17

Point-of-Care Lung Ultrasound in Adults: Image Acquisition

Published on: March 3, 2023

6.1K

Related Experiment Videos

Last Updated: Jul 8, 2025

Protocol and Guidelines for Point-of-Care Lung Ultrasound in Diagnosing Neonatal Pulmonary Diseases Based on International Expert Consensus
06:15

Protocol and Guidelines for Point-of-Care Lung Ultrasound in Diagnosing Neonatal Pulmonary Diseases Based on International Expert Consensus

Published on: March 6, 2019

50.0K
International Expert Consensus and Recommendations for Neonatal Pneumothorax Ultrasound Diagnosis and Ultrasound-guided Thoracentesis Procedure
05:50

International Expert Consensus and Recommendations for Neonatal Pneumothorax Ultrasound Diagnosis and Ultrasound-guided Thoracentesis Procedure

Published on: March 12, 2020

13.7K
Point-of-Care Lung Ultrasound in Adults: Image Acquisition
09:17

Point-of-Care Lung Ultrasound in Adults: Image Acquisition

Published on: March 3, 2023

6.1K

Area of Science:

  • Medical Imaging
  • Artificial Intelligence in Medicine
  • Neonatal Care

Background:

  • Neonatal lung ultrasound (LUS) interpretation is challenging due to similar visual features in normal lungs and Pneumothorax (PTX).
  • Manual M-mode generation for LUS analysis is time-consuming and requires specialized expertise, limiting its clinical application.
  • Automating LUS feature detection can significantly improve diagnostic efficiency in critical care settings.

Purpose of the Study:

  • To develop an interpretable AI system for automated detection of seven common Lung Ultrasound (LUS) features in neonates.
  • To automate M-mode generation from ultrasound videos without human intervention for improved Pneumothorax (PTX) diagnosis.
  • To enhance the accuracy and speed of neonatal lung disease diagnosis through AI-driven LUS analysis.

Main Methods:

  • Utilized object detection models, specifically Faster Region-based Convolutional Neural Network (fRCNN) and RetinaNet, for feature extraction from LUS images.
  • Automated M-mode generation using extracted Regions of Interest (ROIs) and applied a Hough transform-based method for "lung sliding" detection.
  • Evaluated model performance using mean Average Precision (mAP) and accuracy metrics for normal and PTX cases.

Main Results:

  • The fRCNN model achieved a higher mAP (86.57%) compared to RetinaNet (61.15%) at an IoU of 0.2.
  • The system demonstrated high accuracy in classifying ROIs: 97.59% for Normal and 96.37% for PTX videos.
  • Achieved 100% accuracy in classifying 5 PTX and 6 Normal video cases, demonstrating the system's diagnostic capability.

Conclusions:

  • The developed AI system effectively automates the detection of critical LUS features and M-mode generation, addressing limitations of manual interpretation.
  • This automated approach offers a more accurate and efficient method for diagnosing neonatal lung diseases, particularly Pneumothorax (PTX).
  • The system holds significant clinical potential for improving diagnostic workflows and patient outcomes in neonatal intensive care units.