Active, Continual Fine Tuning of Convolutional Neural Networks for Reducing Annotation Efforts
This work addresses the problem of costly and specialized annotation in medical imaging, offering an incremental improvement over existing methods.
The paper tackles the challenge of high annotation costs in medical imaging by integrating active learning and transfer learning into a single framework, demonstrating that it reduces annotation efforts by at least half compared to random selection in three distinct applications.
The splendid success of convolutional neural networks (CNNs) in computer vision is largely attributable to the availability of massive annotated datasets, such as ImageNet and Places. However, in medical imaging, it is challenging to create such large annotated datasets, as annotating medical images is not only tedious, laborious, and time consuming, but it also demands costly, specialty-oriented skills, which are not easily accessible. To dramatically reduce annotation cost, this paper presents a novel method to naturally integrate active learning and transfer learning (fine-tuning) into a single framework, which starts directly with a pre-trained CNN to seek "worthy" samples for annotation and gradually enhances the (fine-tuned) CNN via continual fine-tuning. We have evaluated our method using three distinct medical imaging applications, demonstrating that it can reduce annotation efforts by at least half compared with random selection.