Deep learning uncertainty and confidence calibration for the five-class polyp classification from colonoscopy
Deep learning uncertainty and confidence calibration for the five-class polyp classification from colonoscopy
复制标题
深度学习不确定性和置信度校准用于结肠镜检查的五类息肉分类
DOI：
10.1016/j.media.2020.101653
复制
发表时间：
2020-05-01
影响因子：
10.9
通讯作者：
Burt, Alastair
中科院分区：
文献类型：
--
作者：
Carneiro, Gustavo;Pu, Leonardo Zorron Cheng Tao;Burt, Alastair
在医学图像分析应用中，深度学习模型的可解释性需要解决两个挑战：置信度校准和分类不确定性。置信度校准将分类概率与它实际正确的可能性联系起来——因此，以置信度X%分类的样本有X%的正确分类的机会。分类不确定性估计分类过程中存在的噪声，这种噪声估计可用于评估特定分类结果的可靠性。置信度校准和分类不确定性都被认为有助于解释深度学习模型产生的分类结果，但目前尚不清楚它们对分类精度和校准的影响程度，以及它们如何相互作用。在本文中，我们研究了置信度校准（通过处理后温度缩放）和分类不确定性（从分类熵或贝叶斯方法产生的预测方差计算）在深度学习模型中的作用。结果表明，校正和不确定度提高了分类解释和准确性。这促使我们提出一种新的贝叶斯深度学习方法，该方法同时依赖于校准和不确定性来提高分类精度和模型可解释性。利用940张高质量的结直肠息肉图像数据集，对最近提出的五类息肉分类问题进行了实验，结果表明，我们提出的方法在置信度校准和分类精度方面具有最先进的结果。(C) 2020 Elsevier B.V.版权所有
There are two challenges associated with the interpretability of deep learning models in medical image analysis applications that need to be addressed: confidence calibration and classification uncertainty. Confidence calibration associates the classification probability with the likelihood that it is actually correct - hence, a sample that is classified with confidence X% has a chance of X% of being correctly classified. Classification uncertainty estimates the noise present in the classification process, where such noise estimate can be used to assess the reliability of a particular classification result. Both confidence calibration and classification uncertainty are considered to be helpful in the interpretation of a classification result produced by a deep learning model, but it is unclear how much they affect classification accuracy and calibration, and how they interact. In this paper, we study the roles of confidence calibration (via post-process temperature scaling) and classification uncertainty (computed either from classification entropy or the predicted variance produced by Bayesian methods) in deep learning models. Results suggest that calibration and uncertainty improve classification interpretation and accuracy. This motivates us to propose a new Bayesian deep learning method that relies both on calibration and uncertainty to improve classification accuracy and model interpretability. Experiments are conducted on a recently proposed five-class polyp classification problem, using a data set containing 940 high-quality images of colorectal polyps, and results indicate that our proposed method holds the state-of-the-art results in terms of confidence calibration and classification accuracy. (C) 2020 Elsevier B.V. All rights reserved.