Fine-grained diabetic wound depth and granulation tissue amount assessment using bilinear convolutional neural network.
Fine-grained diabetic wound depth and granulation tissue amount assessment using bilinear convolutional neural network.
复制标题
使用双线性卷积神经网络进行细粒度糖尿病伤口深度和肉芽组织量评估。
DOI：
10.1109/access.2019.2959027
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Kan,Jiangming
中科院分区：
文献类型：
--
作者：
Zhao,Xixuan;Liu,Ziyang;Agu,Emmanuel;Wagh,Ameya;Jain,Shubham;Lindsay,Clifford;Tulu,Bengisu;Strong,Diane;Kan,Jiangming
糖尿病是一种严重的慢性疾病，影响全世界数百万人。在糖尿病患者中，溃疡频繁发生且愈合缓慢。糖尿病溃疡的分级分期是有效治疗的第一步，创面深度和肉芽组织数量是创面愈合进度的两个重要指标。然而，不同严重程度的伤口深度和肉芽组织数量在视觉上可能看起来非常相似，这使得准确的机器学习分类具有挑战性。在本文中，我们创新性地采用了细粒度分类的思想，通过使用双线性CNN（Bi-CNN）架构来处理高度相似的五级图像来进行糖尿病伤口分级。伤口区域提取、锐化、锐化和增强用于在输入到Bi-CNN之前对图像进行预处理。探索通用Bi-CNN网络架构的创新修改以提高其性能。我们的研究产生了一个有价值的伤口数据集。与来自马萨诸塞州医学院的伤口专家合作，我们收集了1639张图像的糖尿病伤口数据集，并用伤口深度和肉芽组织等级作为分类标签对其进行注释。深度学习实验是在这个糖尿病伤口数据集上使用保持验证进行的。与广泛使用的CNN分类架构的比较表明，我们的Bi-CNN细粒度分类方法在糖尿病伤口分级任务方面优于先前的工作。
Diabetes mellitus is a serious chronic disease that affects millions of people worldwide. In patients with diabetes, ulcers occur frequently and heal slowly. Grading and staging of diabetic ulcers is the first step of effective treatment and wound depth and granulation tissue amount are two important indicators of wound healing progress. However, wound depths and granulation tissue amount of different severities can visually appear quite similar, making accurate machine learning classification challenging. In this paper, we innovatively adopted the fine-grained classification idea for diabetic wound grading by using a Bilinear CNN (Bi-CNN) architecture to deal with highly similar images of five grades. Wound area extraction, sharpening, resizing and augmentation were used to pre-process images before being input to the Bi-CNN. Innovative modifications of the generic Bi-CNN network architecture are explored to improve its performance. Our research generated a valuable wound dataset. In collaboration with wound experts from University of Massachusetts Medical School, we collected a diabetic wound dataset of 1639 images and annotated them with wound depth and granulation tissue grades as labels for classification. Deep learning experiments were conducted using holdout validation on this diabetic wound dataset. Comparisons with widely used CNN classification architectures demonstrated that our Bi-CNN fine-grained classification approach outperformed prior work for the task of grading diabetic wounds.