Semantic Segmentation of Smartphone Wound Images: Comparative Analysis of AHRF and CNN-Based Approaches.
Semantic Segmentation of Smartphone Wound Images: Comparative Analysis of AHRF and CNN-Based Approaches.
复制标题
DOI：
10.1109/access.2020.3014175
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Liu Z
中科院分区：
文献类型：
--
作者：
Wagh A;Jain S;Mukherjee A;Agu E;Pedersen P;Strong D;Tulu B;Lindsay C;Liu Z
智能手机伤口图像分析最近已成为一种可行的方法，用于评估愈合进展，并在医院预约之间为患者和护理人员提供可操作的反馈。分割是一个关键的图像分析步骤，之后可以分析伤口段的属性（例如伤口面积和组织成分）。关联分层随机场（AHRF）将图像分割问题归结为一个图优化问题.提取手工制作的特征，然后使用机器学习分类器进行分类。最近，深度学习方法已经出现，并在广泛的图像分析任务中表现出上级性能。FCN、U-Net和DeepLabV 3是用于语义分割的卷积神经网络。虽然在单独的实验中，这些方法中的每一种都显示出有希望的结果，但没有先前的工作在相同的大型伤口图像数据集上全面和系统地比较这些方法，或者更一般地比较深度学习与非深度学习伤口图像分割方法。在本文中，我们使用各种指标比较了AHRF和CNN方法（FCN，U-Net，DeepLabV 3）的分割性能，包括分割精度（骰子得分），推理时间，所需的训练数据量以及不同伤口大小和组织类型的性能。还探讨了使用各种图像预处理和后处理技术可能的改进。由于获得足够的医学图像/数据是一个常见的约束，我们探讨了伤口数据集的大小的方法的灵敏度。我们发现，对于小数据集（< 300张图像），AHRF比U-Net更准确，但不如FCN和DeepLabV 3准确。AHRF也慢了1000倍以上。对于较大的数据集（> 300张图像），AHRF很快饱和，所有CNN方法（FCN，U-Net和DeepLabV 3）都比AHRF更准确。
Smartphone wound image analysis has recently emerged as a viable way to assess healing progress and provide actionable feedback to patients and caregivers between hospital appointments. Segmentation is a key image analysis step, after which attributes of the wound segment (e.g. wound area and tissue composition) can be analyzed. The Associated Hierarchical Random Field (AHRF) formulates the image segmentation problem as a graph optimization problem. Handcrafted features are extracted, which are then classified using machine learning classifiers. More recently deep learning approaches have emerged and demonstrated superior performance for a wide range of image analysis tasks. FCN, U-Net and DeepLabV3 are Convolutional Neural Networks used for semantic segmentation. While in separate experiments each of these methods have shown promising results, no prior work has comprehensively and systematically compared the approaches on the same large wound image dataset, or more generally compared deep learning vs non-deep learning wound image segmentation approaches. In this paper, we compare the segmentation performance of AHRF and CNN approaches (FCN, U-Net, DeepLabV3) using various metrics including segmentation accuracy (dice score), inference time, amount of training data required and performance on diverse wound sizes and tissue types. Improvements possible using various image pre- and post-processing techniques are also explored. As access to adequate medical images/data is a common constraint, we explore the sensitivity of the approaches to the size of the wound dataset. We found that for small datasets (< 300 images), AHRF is more accurate than U-Net but not as accurate as FCN and DeepLabV3. AHRF is also over 1000x slower. For larger datasets (> 300 images), AHRF saturates quickly, and all CNN approaches (FCN, U-Net and DeepLabV3) are significantly more accurate than AHRF.
登录
查看更多内容
影响因子：
3.9
作者：
Beck, Joni;Greenwood, Deborah A.;Wang, Jing
通讯作者：
Wang, Jing
影响因子：
3.1
作者：
Buslaev, Alexander;Iglovikov, Vladimir I.;Kalinin, Alexandr A.
通讯作者：
Kalinin, Alexandr A.
DOI：
10.1109/tpami.2013.165
发表时间：
2014-06-01
影响因子：
23.6
作者：
Ladicky, L'ubor;Russell, Chris;Torr, Philip H. S.
通讯作者：
Torr, Philip H. S.
影响因子：
16.2
作者：
Sheehan, P;Jones, P;Veves, A
通讯作者：
Veves, A
影响因子：
4.8
作者：
DICE, LR
通讯作者：
DICE, LR