Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

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

Carbon-11 Production: Communication, Operations, Maintenance, Troubleshooting, and Analysis for Maintaining High-Grade Bombardment and Provisions of [<sup>11</sup>C]Carbon Dioxide and Its Conversion to [<sup>11</sup>C]Methyl Iodide.

Molecules (Basel, Switzerland)·2026

Same author

Impact of Exposure Parameters on Deep Learning Models in Chest Radiography and Implications for Deployment.

Radiology. Artificial intelligence·2026

Same author

Beyond Water, Rest, and Shade: Advancing Farmworker Heat Protection Through Partnership.

American journal of industrial medicine·2026

Same author

Does Domain-Specific Retrieval Augmented Generation Help LLMs Answer Consumer Health Questions?

Proceedings of machine learning research·2026

Same author

Agentic AI in global health equity for high altitude populations.

NPJ digital medicine·2026

Same author

Impact of Imaging Protocols on Thermal Detection of Pressure Injuries: Threshold versus Deep Learning Across Skin Tones.

medRxiv : the preprint server for health sciences·2026

Search research articles

Related Experiment Video

Updated: Jan 10, 2026

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

Impact of Imaging Protocols on Convolutional Neural Network-Based Pressure Injury Detection.

Miriam Asare-Baiden¹, Sharon Eve Sonenblum², Kathleen Jordan²

¹Emory University, Department of Computer Science and Informatics, Emory University,Atlanta, 30322, USA.

Research Square

|November 24, 2025

Summary

This summary is machine-generated.

Thermal imaging shows high accuracy for early pressure injury detection, outperforming optical imaging. Deep learning models are robust across diverse skin tones and imaging conditions, paving the way for clinical use.

More Related Videos

A Detailed Protocol for Physiological Parameters Acquisition and Analysis in Neurosurgical Critical Patients

A Detailed Protocol for Physiological Parameters Acquisition and Analysis in Neurosurgical Critical Patients

Published on: October 17, 2017

A Bedside, Single Burr Hole Approach to Multimodality Monitoring in Severe Brain Injury

A Bedside, Single Burr Hole Approach to Multimodality Monitoring in Severe Brain Injury

Published on: March 26, 2019

Related Experiment Videos

Last Updated: Jan 10, 2026

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Automated Midline Shift and Intracranial Pressure Estimation based on Brain CT Images

Published on: April 13, 2013

A Detailed Protocol for Physiological Parameters Acquisition and Analysis in Neurosurgical Critical Patients

A Detailed Protocol for Physiological Parameters Acquisition and Analysis in Neurosurgical Critical Patients

Published on: October 17, 2017

A Bedside, Single Burr Hole Approach to Multimodality Monitoring in Severe Brain Injury

A Bedside, Single Burr Hole Approach to Multimodality Monitoring in Severe Brain Injury

Published on: March 26, 2019

Area of Science:

Biomedical Engineering
Medical Imaging
Artificial Intelligence in Healthcare

Background:

Pressure injuries are a significant clinical challenge requiring early detection.
Thermal imaging shows potential for early pressure injury detection, but protocol and skin tone impacts are unclear.

Purpose of the Study:

To assess the influence of imaging protocol variations and patient skin tone on deep learning model performance for early pressure injury detection.
To compare the efficacy of thermal versus optical imaging for this application.

Main Methods:

Collected 1680 images from 35 healthy adults across diverse skin tones under 12 imaging protocols.
Utilized three deep learning models (MobileNetV2, InceptionNetV3, ResNet50) with both optical and thermal imaging.
Simulated temperature changes via localized cooling in a controlled environment.

Main Results:

Thermal imaging achieved >90% accuracy, demonstrating robustness to protocol variations and diverse skin tones.
Optical imaging accuracy varied significantly (32-55%) across protocols, indicating high dependency.
Deep learning models showed sensitivity to both cool and warm regions, suggesting limitations in static labeling.

Conclusions:

Deep learning models trained on thermal imaging data are robust for early pressure injury detection across varied skin tones and imaging conditions.
Thermal imaging is a reliable modality for pressure injury detection, less susceptible to protocol variations than optical imaging.
Findings support the clinical validation of thermal imaging and deep learning for pressure injury prevention.