论文标题
贴花：可部署的临床积极学习
DECAL: DEployable Clinical Active Learning
论文作者
论文摘要
在自然图像上运行的常规机器学习系统假设在图像中存在属性，从而导致某些决定。但是，医疗领域的决策是医疗诊断扫描和电子病历（EMR）中属性的原因。因此，为自然图像开发的主动学习技术不足以处理医疗数据。我们专注于通过在双模式接口中设计可部署的临床活动（贴花）框架来减少这种不足，从而为范式增加实用性。我们的方法是一种“插入”方法，它使基于自然图像的主动学习算法更加更快地推广。我们发现，在两个有关三种体系结构和五种学习策略的医疗数据集上，贴花将20回合的概括提高了约4.81％。作为光学相干断层扫描（OCT）和X射线的初始化策略，贴花的平均准确性增加了5.59％和7.02％。使用3000（5％）和2000（38％）的OCT和X射线数据样本来实现我们的主动学习结果。
Conventional machine learning systems that operate on natural images assume the presence of attributes within the images that lead to some decision. However, decisions in medical domain are a resultant of attributes within medical diagnostic scans and electronic medical records (EMR). Hence, active learning techniques that are developed for natural images are insufficient for handling medical data. We focus on reducing this insufficiency by designing a deployable clinical active learning (DECAL) framework within a bi-modal interface so as to add practicality to the paradigm. Our approach is a "plug-in" method that makes natural image based active learning algorithms generalize better and faster. We find that on two medical datasets on three architectures and five learning strategies, DECAL increases generalization across 20 rounds by approximately 4.81%. DECAL leads to a 5.59% and 7.02% increase in average accuracy as an initialization strategy for optical coherence tomography (OCT) and X-Ray respectively. Our active learning results were achieved using 3000 (5%) and 2000 (38%) samples of OCT and X-Ray data respectively.
