Refinement of an Artificial Intelligence Algorithm for Enhanced Burn Wound Depth Assessment Using Multispectral Imaging: An Expanded Proof of Concept Study
Journal of Burn Care & Research · 05.11.2025 · 46 (6) · 1302-1317
DOI: 10.1093/jbcr/iraf057 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Burn Care & Research, online erschienen am 02.06.2025. Verfasst von Jeffrey E Carter und 6 weiteren Autor:innen. Beteiligte Einrichtung: LSUHSC-New Orleans Department of Surgery, University Medical Center-New Orleans Burn Unit, New Orleans, LA, United States. Hauptthemen laut MEDLINE: Burns, Algorithms, Artificial Intelligence. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 26 Literaturangaben.
Abstract (Originalsprache)
Background
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.
Methods
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).
Results
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.
Conclusions
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.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- LSUHSC-New Orleans Department of Surgery, University Medical Center-New Orleans Burn Unit, New Orleans, LA, United States.
- Medstar Health Burn Center, Washington, DC, United States.
- CMO for Access ProMedical and SiOx Medical, Memphis, TN, United States.
- Cockerell Dermatopathology, Dallas, TX, United States.
- Cardiac Surgery Specialists, Plano, TX, United States.
- Wake Forest School of Medicine Department of Surgery, Atrium Health Wake Forest Baptist Burn Center, Winston Salem, NC, United States.
ORCID-Profile: Herb A Phelan, William Hickerson, Michael DiMaio, James H Holmes
Zitieren
APA
Carter, J. E., Shupp, J. W., Phelan, H. A., Hickerson, W., Cockerell, C. J., DiMaio, M., & Holmes, J. H. (2025). Refinement of an Artificial Intelligence Algorithm for Enhanced Burn Wound Depth Assessment Using Multispectral Imaging: An Expanded Proof of Concept Study. Journal of Burn Care & Research, 46(6), 1302-1317. https://doi.org/10.1093/jbcr/iraf057
Vancouver
Carter JE, Shupp JW, Phelan HA, Hickerson W, Cockerell CJ, DiMaio M, et al. Refinement of an Artificial Intelligence Algorithm for Enhanced Burn Wound Depth Assessment Using Multispectral Imaging: An Expanded Proof of Concept Study. Journal of Burn Care & Research. 2025;46(6):1302-1317. doi:10.1093/jbcr/iraf057
BibTeX
@article{Carter2025,
author = {Carter, Jeffrey E and Shupp, Jeffrey W and Phelan, Herb A and Hickerson, William and Cockerell, Clay J and DiMaio, Michael and Holmes, James H},
title = {Refinement of an Artificial Intelligence Algorithm for Enhanced Burn Wound Depth Assessment Using Multispectral Imaging: An Expanded Proof of Concept Study},
journal = {Journal of Burn Care & Research},
year = {2025},
volume = {46},
number = {6},
pages = {1302--1317},
doi = {10.1093/jbcr/iraf057}
}
EUVIMED