Refinement of an Artificial Intelligence Algorithm for Enhanced Burn Wound Depth Assessment Using Multispectral Imaging: An Expanded Proof of Concept Study

With the advent of convolutional neural networks (CNNs), artificial intelligence is now applicable to visual fields. We used multispectral imaging (MSI) sensors capable of detecting wavelengths outside visible spectra to image burn wounds. The output was converted to pixel-level data and analyzed by...

Full description

Saved in:
Bibliographic Details
Published in:Journal of burn care & research Vol. 46; no. 6; p. 1302
Main Authors: Carter, Jeffrey E, Shupp, Jeffrey W, Phelan, Herb A, Hickerson, William, Cockerell, Clay J, DiMaio, Michael, Holmes, 4th, James H
Format: Journal Article
Language:English
Published: England 05.11.2025
Subjects:
Adult
Algorithms
Artificial Intelligence
Burns - diagnostic imaging
Burns - pathology
Burns - therapy
Child
Deep Learning
Female
Humans
Male
Middle Aged
Neural Networks, Computer
Proof of Concept Study
Prospective Studies
burn depth
deep learning
convolutional neural network
artificial intelligence
ISSN:1559-0488, 1559-0488
Online Access:Get more information
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:With the advent of convolutional neural networks (CNNs), artificial intelligence is now applicable to visual fields. We used multispectral imaging (MSI) sensors capable of detecting wavelengths outside visible spectra to image burn wounds. The output was converted to pixel-level data and analyzed by an array of CNNs to inform the development of a deep learning (DL) algorithm for burn assessment. Three burn centers prospectively grouped consenting subjects into those with wounds likely to heal nonoperatively by 21 days, or those benefiting from surgery. Both groups underwent MSI sensor imaging at enrollment and once daily until discharge/excision. Nonoperative subjects were evaluated at 21 days, while operative subjects underwent biopsies. A "Truthing Panel" of burn experts created a "ground truth" for each wound that was converted to pixel-level data and used to train ten CNNs (8 unique DL algorithms and 2 ensemble DL algorithms). 1037 MSI images and 161 biopsies were collected from 100 adult and 24 pediatric subjects. The most effective CNN algorithm exhibited an area under the curve of 0.95 (accuracy = 89.29%, sensitivity = 90.51%, and specificity = 87.22%) with the covariate "time-since-injury" found to be significant (P < .0001). Accuracy was lowest, 88.5%, at 1-2 days after injury and highest, 93.5%, at 3-4 days. The CNN's learning curve predicted an accuracy of 94.04% after enrolling 374 subjects in a future training study. An optimal CNN architecture and the importance of "time-since-injury" as a covariate were identified, informing the design/powering of upcoming algorithm training and validation studies.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1559-0488
1559-0488
DOI:10.1093/jbcr/iraf057