Evaluation Framework for Bruise Detection: Systematic ALS/White-Light Training and Skin-Tone Balancing with Deep Learning
Sensors · 19.05.2026 · 26 (10) · 3215
DOI: 10.3390/s26103215 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 19.05.2026. Verfasst von Kiyarash Aminfar und 3 weiteren Autor:innen. Beteiligte Einrichtung: Sid and Reva Dewberry Department of Civil, Environmental, and Infrastructure Engineering, College of Engineering and Computing, George Mason Universi…. Hauptthemen laut MEDLINE: Deep Learning, Contusions, Skin Pigmentation. Die Arbeit nennt 62 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science Foundation, National Institute of Justice, Office of Research Computing at George Mason University.
Abstract (Originalsprache)
Accurate and consistent forensic bruise assessment is critical in ensuring positive clinical and legal outcomes for victims of violence. In this study, a framework for automated bruise detection is presented that, for the first time, integrates narrowband alternate-light-source (ALS) forensic imaging and ambient white light imaging. This evaluation framework is designed to address long-standing issues with respect to equitable performance across skin tones and lighting scenarios via a combination of novel model diagnostic strategies. In particular, skin-tone balancing during training and testing, threshold-sensitivity analysis, and embedding-similarity partitioning are employed to quantify the model robustness and deployment trade-offs that arise in forensic image analysis. Models were implemented with ImageNet-pretrained backbones and trained on a unique, multi-annotator full-consensus dataset comprising both white-light and ALS (415 nm and 450 nm) images. The protocol emphasizes three axes of operational relevance: (1) illumination composition in training (W/ALS ratio); (2) subgroup fairness via targeted balancing; and (3) model operating-point selection (confidence and IoU thresholds) informed by confidence-stability metrics and bootstrapped uncertainty estimates. Systematic W/ALS ratio sweeps indicate peak accuracy under ALS-dominant training and declining performance as the proportion of white-light images increases within the training set. Skin-tone balancing reduced failure rates for darker skin tones but increased overprediction in some demographic subgroups. Embedding-similarity and seen/unseen injury analyses demonstrate inflated generalization under image-level partitioning. Ultimately, the findings suggest that future researchers and developers should employ injury-level data partitioning and ensure a weighted balance of ALS images during training.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Sid and Reva Dewberry Department of Civil, Environmental, and Infrastructure Engineering, College of Engineering and Computing, George Mason University, Fairfax, VA 22030, USA.
- School of Nursing, College of Public Health, George Mason University, Fairfax, VA 22030, USA.
- Department of Global and Community Health, College of Public Health, George Mason University, Fairfax, VA 22030, USA.
ORCID-Profile: Kiyarash Aminfar, David Lattanzi
Förderung
- National Science Foundation
- National Institute of Justice
- Office of Research Computing at George Mason University
Zitieren
APA
Aminfar, K., Scafide, K., Wojtusiak, J., & Lattanzi, D. (2026). Evaluation Framework for Bruise Detection: Systematic ALS/White-Light Training and Skin-Tone Balancing with Deep Learning. Sensors, 26(10), 3215. https://doi.org/10.3390/s26103215
Vancouver
Aminfar K, Scafide K, Wojtusiak J, Lattanzi D. Evaluation Framework for Bruise Detection: Systematic ALS/White-Light Training and Skin-Tone Balancing with Deep Learning. Sensors. 2026;26(10):3215. doi:10.3390/s26103215
BibTeX
@article{Aminfar2026,
author = {Aminfar, Kiyarash and Scafide, Katherine and Wojtusiak, Janusz and Lattanzi, David},
title = {Evaluation Framework for Bruise Detection: Systematic ALS/White-Light Training and Skin-Tone Balancing with Deep Learning},
journal = {Sensors},
year = {2026},
volume = {26},
number = {10},
pages = {3215},
doi = {10.3390/s26103215}
}
EUVIMED