Automated Wound Image Segmentation: Transfer Learning from Human to Pet via Active Semi-Supervised Learning

Abstract

Wound management is a fundamental task in standard clinical practice. Automated solutions already exist for humans, but there is a lack of applications regarding wound management for pets. Precise and efficient wound assessment is helpful to improve diagnosis and to increase the effectiveness of treatment plans for chronic wounds. In this work, we introduced a novel pipeline for the segmentation of pet wound images. Starting from a model pre-trained on human-based wound images, we applied a combination of transfer learning (TL) and active semi-supervised learning (ASSL) to automatically label a large dataset. Additionally, we provided a guideline for future applications of TL+ASSL training strategy on image datasets. We compared the effectiveness of the proposed training strategy, monitoring the performance of an EfficientNet-b3 U-Net model against the lighter solution provided by a MobileNet-v2 U-Net model. We obtained 80% of correctly segmented images after five rounds of ASSL training. The EfficientNet-b3 U-Net model significantly outperformed the MobileNet-v2 one. We proved that the number of available samples is a key factor for the correct usage of ASSL training. The proposed approach is a viable solution to reduce the time required for the generation of a segmentation dataset.

Automated wound care by employing a reliable U-Net architecture combined with ResNet feature encoders for monitoring chronic wounds

“…The utilization of U-Net has been widely observed in wound segmentation. Recently, numerous studies (5,19,30,(45)(46)(47) have endeavored to improve their methodologies by developing enhanced models that are built upon the U-Net framework. This study presents a novel framework that combines the U-Net architecture and the ResNet34 model to improve the effectiveness of image segmentation tasks.…”

“…The utilization of the existing AZH Woundcare and Medetec datasets has resulted in the establishment of a novel state-of-the-art. Buschi et al ( 30 ) proposed a methodology to segment the pet wound images automatically. This approach involves the utilization of transfer learning (TL) and active self-supervised learning (ASSL) techniques.…”