Error Corrective Boosting for Learning Fully Convolutional Networks with Limited Data
This work addresses the challenge of limited labeled data in medical image segmentation, particularly for brain MRI, by proposing a method that improves efficiency and accuracy, though it is incremental as it builds on existing F-CNN techniques.
The paper tackles the problem of training fully convolutional networks for semantic segmentation with limited labeled data by using automatically generated auxiliary labels and a new error corrective boosting method, achieving a segmentation speed of 7 seconds per 3D scan compared to 30 hours for a multi-atlas method while maintaining similar performance.
Training deep fully convolutional neural networks (F-CNNs) for semantic image segmentation requires access to abundant labeled data. While large datasets of unlabeled image data are available in medical applications, access to manually labeled data is very limited. We propose to automatically create auxiliary labels on initially unlabeled data with existing tools and to use them for pre-training. For the subsequent fine-tuning of the network with manually labeled data, we introduce error corrective boosting (ECB), which emphasizes parameter updates on classes with lower accuracy. Furthermore, we introduce SkipDeconv-Net (SD-Net), a new F-CNN architecture for brain segmentation that combines skip connections with the unpooling strategy for upsampling. The SD-Net addresses challenges of severe class imbalance and errors along boundaries. With application to whole-brain MRI T1 scan segmentation, we generate auxiliary labels on a large dataset with FreeSurfer and fine-tune on two datasets with manual annotations. Our results show that the inclusion of auxiliary labels and ECB yields significant improvements. SD-Net segments a 3D scan in 7 secs in comparison to 30 hours for the closest multi-atlas segmentation method, while reaching similar performance. It also outperforms the latest state-of-the-art F-CNN models.