Pressure Ulcer Detection and Classification Using Convolutional Neural Networks: Development of a Mobile Clinical Support Tool
Year: 2026
Type: article
Abstract: The aim of this study is to develop a mobile application for the automated detection and classification of pressure ulcers (PU) using convolutional neural networks (CNNs). The methodology of this study was developed using a deep detection approach for the identification and classification of PU from clinical imaging. CNN methods were used to classify PU images according to their degree of damage. To evaluate its effectiveness, various CNNs architectures were compared and the model with the best classification accuracy was selected. The results showed that YOLOv11-CLS managed to achieve the highest overall comparative accuracy at 97.1%. Regarding the PU detection task, the model obtained a recall of 96.1%, a precision of 98.3%, an F1-score of 97.2%, and an accuracy of 98.1%. Likewise, for the multiclass severity classification, it managed to achieve a recall of 95.7%, a precision of 97.6%, an F1-score of 96.9%, and an accuracy of 97.9%, which guarantees a high efficiency of the application. In conclusion, the proposed mobile system represents a promising tool for the early detection of PU and can be used as a complement to improve the effectiveness of initial medical follow-up.
Institution: Universidad de Ciencias y Humanidades
Language: English
Subfield: Dermatology
Field: Medicine
Domain: Health Sciences
Open Access status: gold
Estimated APC paid: $312
OpenAlex ID: W7222396576