TumorCP: A Simple but Effective Object-Level Data Augmentation for Tumor Segmentation

TL;DR

TumorCP is a novel object-level data augmentation technique for tumor segmentation that significantly improves model performance, especially in low-data scenarios, by providing diverse and unlimited tumor augmentations.

Contribution

This work introduces TumorCP, the first extension of 'Copy-Paste' augmentation to medical imaging, enhancing tumor segmentation accuracy and data efficiency.

Findings

TumorCP improves tumor Dice by 7.12% over baseline.

Combining TumorCP with image augmentation surpasses state-of-the-art by 2.32%.

In low-data settings, TumorCP boosts tumor Dice by 21.87%.

Abstract

Deep learning models are notoriously data-hungry. Thus, there is an urging need for data-efficient techniques in medical image analysis, where well-annotated data are costly and time consuming to collect. Motivated by the recently revived "Copy-Paste" augmentation, we propose TumorCP, a simple but effective object-level data augmentation method tailored for tumor segmentation. TumorCP is online and stochastic, providing unlimited augmentation possibilities for tumors' subjects, locations, appearances, as well as morphologies. Experiments on kidney tumor segmentation task demonstrate that TumorCP surpasses the strong baseline by a remarkable margin of 7.12% on tumor Dice. Moreover, together with image-level data augmentation, it beats the current state-of-the-art by 2.32% on tumor Dice. Comprehensive ablation studies are performed to validate the effectiveness of TumorCP. Meanwhile, we show that TumorCP can lead to striking improvements in extremely low-data regimes. Evaluated with only 10% labeled data, TumorCP significantly boosts tumor Dice by 21.87%. To the best of our knowledge, this is the very first work exploring and extending the "Copy-Paste" design in medical imaging domain. Code is available at: https://github.com/YaoZhang93/TumorCP.