Explainable, Multi-modal Wound Infection Classification from Images Augmented with Generated Captions
ACM Transactions on Computing for Healthcare · 30.04.2026 · 7 (2) · 1-28
DOI: 10.1145/3793535 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Computing for Healthcare, online erschienen am 17.03.2026. Verfasst von Palawat Busaranuvong und 7 weiteren Autor:innen. Die Arbeit nennt 65 Literaturangaben.
Abstract (Originalsprache)
Infections in Diabetic Foot Ulcers (DFUs) can cause severe complications, including tissue death and limb amputation, highlighting the need for accurate, timely diagnosis. Previous machine learning methods have focused on identifying infections by analyzing wound images alone, without utilizing additional metadata such as medical notes. In this study, we aim to improve infection classification by introducing S ynthetic C aption A ugmented R etrieval for W ound I nfection D etection (SCARWID), a novel multimodal deep learning framework that leverages synthetic textual descriptions to augment DFU images. SCARWID consists of two components: (1) Wound-BLIP, a Vision-Language Model (VLM) fine-tuned on GPT-4o-generated descriptions to synthesize consistent captions from images; and (2) an Image-Text Fusion module that uses cross-attention to extract cross-modal embeddings from an image and its corresponding Wound-BLIP caption. Infection status is determined by retrieving the top- \(k\) similar items from a labeled support set. To enhance the diversity of training data, we utilized a latent diffusion model to generate additional wound images. As a result, SCARWID outperformed state-of-the-art models, achieving average accuracy, sensitivity, and specificity of 0.814, 0.845, and 0.783, respectively, for wound infection classification.
Zitieren
APA
Busaranuvong, P., Agu, E., Fard, R. S., Kumar, D., Gautam, S., Tulu, B., Strong, D., & Loretz, L. (2026). Explainable, Multi-modal Wound Infection Classification from Images Augmented with Generated Captions. ACM Transactions on Computing for Healthcare, 7(2), 1-28. https://doi.org/10.1145/3793535
Vancouver
Busaranuvong P, Agu E, Fard RS, Kumar D, Gautam S, Tulu B, et al. Explainable, Multi-modal Wound Infection Classification from Images Augmented with Generated Captions. ACM Transactions on Computing for Healthcare. 2026;7(2):1-28. doi:10.1145/3793535
BibTeX
@article{Busaranuvong2026,
author = {Busaranuvong, Palawat and Agu, Emmanuel and Fard, Reza Saadati and Kumar, Deepak and Gautam, Shefalika and Tulu, Bengisu and Strong, Diane and Loretz, Lorraine},
title = {Explainable, Multi-modal Wound Infection Classification from Images Augmented with Generated Captions},
journal = {ACM Transactions on Computing for Healthcare},
year = {2026},
volume = {7},
number = {2},
pages = {1--28},
doi = {10.1145/3793535}
}
EUVIMED