Neuwieser Hannah, Jami Naga Venkata Sai Jitin, Meier Robert Johannes, Liebsch Gregor, Felthaus Oliver, Klein Silvan, Schreml Stephan, Berneburg Mark, Prantl Lukas, Leutheuser Heike, Kempa Sally
Department of Plastic, Hand, and Reconstructive Surgery, University Hospital Regensburg, 93053 Regensburg, Germany.
Ambient Assisted Living & Medical Assistance Systems, Department of Computer Science, University of Bayreuth, 95447 Bayreuth, Germany.
Diagnostics (Basel). 2025 Aug 28;15(17):2184. doi: 10.3390/diagnostics15172184.
: Chronic wounds of the lower extremities, particularly arterial and venous ulcers, represent a significant and costly challenge in medical care. To assist in differential diagnosis, we aim to evaluate various advanced deep-learning models for classifying arterial and venous ulcers and visualize their decision-making processes. : A retrospective dataset of 607 images (198 arterial and 409 venous ulcers) was used to train five convolutional neural networks: ResNet50, ResNeXt50, ConvNeXt, EfficientNetB2, and EfficientNetV2. Model performance was assessed using accuracy, precision, recall, F1-score, and ROC-AUC. Grad-CAM was applied to visualize image regions contributing to classification decisions. : The models demonstrated high classification performance, with accuracy ranging from 72% (ConvNeXt) to 98% (ResNeXt50). Precision and recall values indicated strong discrimination between arterial and venous ulcers, with EfficientNetV2 achieving the highest precision. : AI-assisted classification of venous and arterial ulcers offers a valuable method for enhancing diagnostic efficiency.
下肢慢性伤口，尤其是动脉性和静脉性溃疡，是医疗保健中一项重大且成本高昂的挑战。为了协助进行鉴别诊断，我们旨在评估各种先进的深度学习模型，以对动脉性和静脉性溃疡进行分类，并可视化它们的决策过程。：使用一个包含607张图像（198张动脉性溃疡和409张静脉性溃疡）的回顾性数据集来训练五个卷积神经网络：ResNet50、ResNeXt50、ConvNeXt、EfficientNetB2和EfficientNetV2。使用准确率、精确率、召回率、F1分数和ROC-AUC来评估模型性能。应用Grad-CAM来可视化对分类决策有贡献的图像区域。：这些模型表现出了较高的分类性能，准确率范围从72%（ConvNeXt）到98%（ResNeXt50）。精确率和召回率值表明动脉性和静脉性溃疡之间有很强的区分度，EfficientNetV2的精确率最高。：人工智能辅助的静脉性和动脉性溃疡分类为提高诊断效率提供了一种有价值的方法。
