Effectiveness of Semi-Supervised Active Learning in Automated Wound Image Segmentation

Abstract

Appropriate wound management shortens the healing times and reduces the management costs, benefiting the patient in physical terms and potentially reducing the healthcare system’s economic burden. Among the instrumental measurement methods, the image analysis of a wound area is becoming one of the cornerstones of chronic ulcer management. Our study aim is to develop a solid AI method based on a convolutional neural network to segment the wounds efficiently to make the work of the physician more efficient, and subsequently, to lay the foundations for the further development of more in-depth analyses of ulcer characteristics. In this work, we introduce a fully automated model for identifying and segmenting wound areas which can completely automatize the clinical wound severity assessment starting from images acquired from smartphones. This method is based on an active semi-supervised learning training of a convolutional neural network model. In our work, we tested the robustness of our method against a wide range of natural images acquired in different light conditions and image expositions. We collected the images using an ad hoc developed app and saved them in a database which we then used for AI training. We then tested different CNN architectures to develop a balanced model, which we finally validated with a public dataset. We used a dataset of images acquired during clinical practice and built an annotated wound image dataset consisting of 1564 ulcer images from 474 patients. Only a small part of this large amount of data was manually annotated by experts (ground truth). A multi-step, active, semi-supervised training procedure was applied to improve the segmentation performances of the model. The developed training strategy mimics a continuous learning approach and provides a viable alternative for further medical applications. We tested the efficiency of our model against other public datasets, proving its robustness. The efficiency of the transfer learning showed that after less than 50 epochs, the model achieved a stable DSC that was greater than 0.95. The proposed active semi-supervised learning strategy could allow us to obtain an efficient segmentation method, thereby facilitating the work of the clinician by reducing their working times to achieve the measurements. Finally, the robustness of our pipeline confirms its possible usage in clinical practice as a reliable decision support system for clinicians.

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

“…In this work, we developed an ad hoc Active Semi-Supervised Learning (ASSL) strategy, reproducing the procedure already discussed in our previous work [ 23 ]. The ASSL procedure was reiterated for several rounds of training, requiring the clinicians to simply accept or discard the output of the model after each round.…”

“…The validated images (and corresponding segmentation masks) which satisfied all the criteria were inserted into the training set as the starting point of the current ASSL training strategy. Ad hoc software was developed to minimize the time required for the validation procedure [ 29 ]. The validation was performed, reviewing the segmentations overlayed on the original image displayed on a high-resolution screen and without limits of time for the evaluation.…”

“…Another interesting attempt by Mahapatra et al [ 22 ] introduced the usage of Semi-Supervised Learning (SSL) in combination with Active Learning (AL) on Crohn’s disease; starting from only a few labelled samples, the predictions of the automated model were iteratively refined by experts in different rounds of training. Following the same training strategy, we discussed in our previous work [ 23 ] an application to human wound image segmentation using a deep learning model, obtaining state-of-the-art comparable results.…”

“…We aim to minimize the labeling effort of the clinicians, requiring no starting manual annotation at all. To perform a robust image segmentation without a labelled ground truth, we started from the model that Curti et al [ 23 ] trained on a human wound image dataset.…”

Recent Approaches for Wound Treatment

“…Curti et al [ 2 ] suggested an automated model able to perform a deep analysis of wounds from images acquired from smartphones using an app developed ad hoc . This method is based on an active semi-supervised learning training of a convolutional neural network.…”