Dual Meta-Learning with Longitudinally Generalized Regularization for One-Shot Brain Tissue Segmentation Across the Human Lifespan
This work addresses a domain-specific problem in neuroscience and clinical studies by improving segmentation accuracy for longitudinal brain data, though it appears incremental as it builds on existing meta-learning and regularization approaches.
The paper tackles the challenge of brain tissue segmentation across the human lifespan by proposing a dual meta-learning paradigm with regularizations to maintain longitudinal consistency, achieving effective results on iSeg2019 and ADNI datasets.
Brain tissue segmentation is essential for neuroscience and clinical studies. However, segmentation on longitudinal data is challenging due to dynamic brain changes across the lifespan. Previous researches mainly focus on self-supervision with regularizations and will lose longitudinal generalization when fine-tuning on a specific age group. In this paper, we propose a dual meta-learning paradigm to learn longitudinally consistent representations and persist when fine-tuning. Specifically, we learn a plug-and-play feature extractor to extract longitudinal-consistent anatomical representations by meta-feature learning and a well-initialized task head for fine-tuning by meta-initialization learning. Besides, two class-aware regularizations are proposed to encourage longitudinal consistency. Experimental results on the iSeg2019 and ADNI datasets demonstrate the effectiveness of our method. Our code is available at https://github.com/ladderlab-xjtu/DuMeta.