Dual Meta-Learning with Longitudinally Generalized Regularization for One-Shot Brain Tissue Segmentation Across the Human Lifespan
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 25 of its 42 references that are in the map. First posted on arXiv 13 Aug 2023.
Builds on
25 of 42 references.
References · 25
- 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
- 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
- 2020Exploring Simple Siamese Representation Learningcited by 2,866
- 2016V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentationcited by 2,292
- 2018On First-Order Meta-Learning Algorithmscited by 1,018
- 2020Meta-Learning in Neural Networks: A Surveycited by 950
- 2020Contrastive Learning for Unpaired Image-to-Image Translationcited by 639
- 2018Bilevel Programming for Hyperparameter Optimization and Meta-Learningcited by 405
- 2017Meta-SGD: Learning to Learn Quickly for Few-Shot Learningcited by 379
- 2019Rapid Learning or Feature Reuse? Towards Understanding the Effectiveness of MAMLcited by 359
- 2020Contrastive learning of global and local features for medical image segmentation with limited annotationscited by 309
- 2021Investigating Bi-Level Optimization for Learning and Vision from a Unified Perspective: A Survey and Beyondcited by 153
13 more
- 2020Shape-aware Meta-learning for Generalizing Prostate MRI Segmentation to Unseen Domainscited by 124
- 2018Stochastic Hyperparameter Optimization through Hypernetworkscited by 73
- 2019ES-MAML: Simple Hessian-Free Meta Learningcited by 57
- 2021Positional Contrastive Learning for Volumetric Medical Image Segmentationcited by 56
- 2021Semi-supervised Meta-learning with Disentanglement for Domain-generalised Medical Image Segmentationcited by 36
- 2020Multi-Site Infant Brain Segmentation Algorithms: The iSeg-2019 Challengecited by 28
- 2022Local Spatiotemporal Representation Learning for Longitudinally-consistent Neuroimage Analysiscited by 19
- 2015Bilevel approaches for learning of variational imaging modelscited by 17
- 2021Self-Supervised Longitudinal Neighbourhood Embeddingcited by 16
- 2014Automatic Differentiation of Algorithms for Machine Learningcited by 7
- 2020Longitudinal diffusion MRI analysis using Segis-Net: a single-step deep-learning framework for simultaneous segmentation and registrationcited by 7
- 2017A Bridge Between Hyperparameter Optimization and Learning-to-learncited by 6
- 2023Brain Tissue Segmentation Across the Human Lifespan via Supervised Contrastive Learningcited by 4
Built on by
Citing papers per year: 2025: 1; 2026: 1.
2 papers.