Computer-aided diagnosis from weak supervision: A benchmarking study
Computer-aided diagnosis from weak supervision: A benchmarking study
复制标题
DOI：
10.1016/j.compmedimag.2014.11.010
复制
发表时间：
2015-06-01
影响因子：
5.7
通讯作者：
Hamprecht, Fred A.
中科院分区：
文献类型：
--
作者：
Kandemir, Melih;Hamprecht, Fred A.
监督机器学习是计算机辅助诊断（CAD）应用中经常使用的强大工具。该技术的瓶颈是它需要细粒度的专家注释，这对于医学图像分析应用来说是繁琐的。此外，信息通常被定位在诊断图像中，这使得通过单个特征集来表示整个图像成为问题。多实例学习框架作为这两个问题的补救措施，允许为观察组（称为包）提供标签，并假设组标签是包内实例标签的最大值。通过将给定的诊断图像分割成笛卡尔网格，通过用特征集表示每个网格元素（补丁）来将其视为实例，并将属于同一图像的实例分组到袋中，可以将该设置有效地应用于CAD。我们通过评估现有多实例学习方法在两个不同CAD应用程序上的性能来量化其功能：（i）Barrett癌症诊断和（ii）糖尿病视网膜病变筛查。在实验中，对于这两种具有截然不同视觉特征的应用程序，mi-Graph似乎是袋级预测（即诊断）中表现最好的方法。对于实例级预测（即疾病定位），mi-SVM是最准确的方法。(C)2014爱思唯尔有限公司版权所有。
Supervised machine learning is a powerful tool frequently used in computer-aided diagnosis (CAD) applications. The bottleneck of this technique is its demand for fine grained expert annotations, which are tedious for medical image analysis applications. Furthermore, information is typically localized in diagnostic images, which makes representation of an entire image by a single feature set problematic. The multiple instance learning framework serves as a remedy to these two problems by allowing labels to be provided for groups of observations, called bags, and assuming the group label to be the maximum of the instance labels within the bag. This setup can effectively be applied to CAD by splitting a given diagnostic image into a Cartesian grid, treating each grid element (patch) as an instance by representing it with a feature set, and grouping instances belonging to the same image into a bag. We quantify the power of existing multiple instance learning methods by evaluating their performance on two distinct CAD applications: (i) Barrett's cancer diagnosis and (ii) diabetic retinopathy screening. In the experiments, mi-Graph appears as the best-performing method in bag-level prediction (i.e. diagnosis) for both of these applications that have drastically different visual characteristics. For instance-level prediction (i.e. disease localization), mi-SVM ranks as the most accurate method. (C) 2014 Elsevier Ltd. All rights reserved.