论文标题
胸部X射线分类中的域适应性持续学习
Continual Learning for Domain Adaptation in Chest X-ray Classification
论文作者
论文摘要
在过去的几年中，深度学习已成功应用于广泛的医疗应用。尤其是在胸部X射线分类的情况下，据报道，结果在PAR上，甚至优于经验丰富的放射科医生。尽管在受控的实验环境中取得了成功，但已经注意到，深度学习模型从新领域（具有不同任务）的数据推广到数据的能力通常受到限制。为了应对这一挑战，我们研究了持续学习（CL）领域的技术，包括联合培训（JT），弹性重量巩固（EWC）和学习而不忘记（LWF）。使用ChestX-Ray14和Mimic-CXR数据集，我们从经验上证明，这些方法为提高目标域上深度学习模型的性能提供了有希望的选择，并有效地减轻了对源域的灾难性遗忘。为此，最佳的总体性能是使用JT获得的，而对于LWF竞争结果，即使不访问源域中的数据也可以实现。
Over the last years, Deep Learning has been successfully applied to a broad range of medical applications. Especially in the context of chest X-ray classification, results have been reported which are on par, or even superior to experienced radiologists. Despite this success in controlled experimental environments, it has been noted that the ability of Deep Learning models to generalize to data from a new domain (with potentially different tasks) is often limited. In order to address this challenge, we investigate techniques from the field of Continual Learning (CL) including Joint Training (JT), Elastic Weight Consolidation (EWC) and Learning Without Forgetting (LWF). Using the ChestX-ray14 and the MIMIC-CXR datasets, we demonstrate empirically that these methods provide promising options to improve the performance of Deep Learning models on a target domain and to mitigate effectively catastrophic forgetting for the source domain. To this end, the best overall performance was obtained using JT, while for LWF competitive results could be achieved - even without accessing data from the source domain.
