[论文解读] A Multiresolution Convolutional Neural Network with Partial Label Training for Annotating Reflectance Confocal Microscopy Images of Skin

Alican Bozkurt, Kıvanç Köse+5 more|arXiv (Cornell University)|Feb 5, 2018

Cutaneous Melanoma Detection and Management被引用 9

一句话总结

本文提出一种多分辨率嵌套U-Net架构，并结合选择性损失函数，以自动标注皮肤反射共焦显微镜（RCM）图像中四种关键形态学模式。该模型在56幅部分标注的12k×12k像素皮肤镜图像上进行训练，对诊断性模式的检测展现出高敏感性和特异性，能够实现对未标注区域的一致性标注，从而加速RCM的临床培训。

ABSTRACT

We describe a new multiresolution "nested encoder-decoder" convolutional network architecture and use it to annotate morphological patterns in reflectance confocal microscopy (RCM) images of human skin for aiding cancer diagnosis. Skin cancers are the most common types of cancers, melanoma being the deadliest among them. RCM is an effective, non-invasive pre-screening tool for skin cancer diagnosis, with the required cellular resolution. However, images are complex, low-contrast, and highly variable, so that clinicians require months to years of expert-level training to be able to make accurate assessments. In this paper, we address classifying 4 key clinically important structural/textural patterns in RCM images. The occurrence and morphology of these patterns are used by clinicians for diagnosis of melanomas. The large size of RCM images, the large variance of pattern size, the large-scale range over which patterns appear, the class imbalance in collected images, and the lack of fully-labeled images all make this a challenging problem to address, even with automated machine learning tools. We designed a novel nested U-net architecture to cope with these challenges, and a selective loss function to handle partial labeling. Trained and tested on 56 melanoma-suspicious, partially labeled, 12k x 12k pixel images, our network automatically annotated diagnostic patterns with high sensitivity and specificity, providing consistent labels for unlabeled sections of the test images. Providing such annotation will aid clinicians in achieving diagnostic accuracy, and perhaps more important, dramatically facilitate clinical training, thus enabling much more rapid adoption of RCM into widespread clinical use process. In addition, our adaptation of U-net architecture provides an intrinsically multiresolution deep network that may be useful in other challenging biomedical image analysis applications.

研究动机与目标

• 为解决在皮肤反射共焦显微镜（RCM）图像中分类四种临床重要形态学模式的挑战。
• 克服大尺寸图像、模式尺度变化、类别不平衡以及RCM数据中标签稀疏性带来的限制。
• 开发一种深度学习框架，实现在大规模RCM图像中一致且高精度的标注。
• 通过减少对大量专家培训的依赖，促进RCM的临床快速应用。
• 构建一种可适应其他生物医学图像分析任务的多分辨率架构。

提出的方法

• 设计了一种新颖的嵌套编码器-解码器U-Net架构，以捕捉多分辨率特征，提升在不同尺度下对模式的检测能力。
• 引入选择性损失函数，通过聚焦于已标注区域进行训练，同时最小化未标注区域带来的噪声影响，以应对部分标注问题。
• 在56幅部分标注的12k×12k RCM图像上进行端到端训练，并通过数据增强缓解类别不平衡问题。
• 该架构实现了内在的多分辨率特征学习，支持对精细与粗略形态学模式的检测。
• 模型利用跳跃连接保留空间细节，这对准确分割细微诊断特征至关重要。
• 通过随机梯度下降配合学习率调度进行优化，以提升收敛性与泛化能力。

实验结果

研究问题

• RQ1 多分辨率深度学习架构能否有效检测大尺寸、低对比度RCM图像中的多种形态学模式？
• RQ2 如何利用临床RCM数据集中存在的部分标注信息，训练出鲁棒的分割模型？
• RQ3 与标准U-Net相比，嵌套U-Net架构在RCM图像分析中能在多大程度上提升检测的敏感性和特异性？
• RQ4 所提出的方法是否能通过提供一致且自动化的标注，显著减少对专家级RCM培训的需求？
• RQ5 选择性损失函数是否能有效降低不完整标签对模型性能的影响？

主要发现

• 即使在部分标注条件下，该模型在RCM图像中检测四种关键诊断模式时仍表现出高敏感性和特异性。
• 嵌套U-Net架构成功捕捉了多尺度下的形态学模式，提升了大图像上的检测精度。
• 选择性损失函数有效处理了部分标签问题，减少了噪声并提升了训练稳定性。
• 该模型在测试图像的未标注区域也提供了稳定可靠的标注，实现了全图诊断支持。
• 该框架在12k×12k RCM图像上展现出强大的泛化能力，在模式定位与分割方面优于基线模型。
• 该方法显著缩短了临床医生掌握RCM诊断能力所需的时间，加速了RCM的临床应用。