DuMeta++: Spatiotemporal Dual Meta-Learning for Generalizable Few-Shot Brain Tissue Segmentation Across Diverse Ages

2026-02-06

AI Summary

This work addresses the challenge of degraded generalization performance in cross-age brain tissue segmentation from MRI due to dynamic anatomical changes across the lifespan. To tackle this issue without paired longitudinal data, the authors propose DuMeta++, a novel framework that jointly optimizes meta-feature extraction and meta-initialization through a dual learning mechanism. It further incorporates a memory bank–based, class-aware regularization strategy to enforce temporal consistency in segmentation without explicit temporal supervision. Theoretical analysis establishes the convergence of the proposed algorithm. Extensive few-shot cross-age experiments on multiple datasets—including iSeg-2019, IBIS, OASIS, and ADNI—demonstrate that DuMeta++ significantly outperforms existing methods, achieving both high segmentation accuracy and strong generalization capability across diverse age groups.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Metareasoning and Metaheuristics
• Computer Vision: Segmentation

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Accurate segmentation of brain tissues from MRI scans is critical for neuroscience and clinical applications, but achieving consistent performance across the human lifespan remains challenging due to dynamic, age-related changes in brain appearance and morphology. While prior work has sought to mitigate these shifts by using self-supervised regularization with paired longitudinal data, such data are often unavailable in practice. To address this, we propose \emph{DuMeta++}, a dual meta-learning framework that operates without paired longitudinal data. Our approach integrates: (1) meta-feature learning to extract age-agnostic semantic representations of spatiotemporally evolving brain structures, and (2) meta-initialization learning to enable data-efficient adaptation of the segmentation model. Furthermore, we propose a memory-bank-based class-aware regularization strategy to enforce longitudinal consistency without explicit longitudinal supervision. We theoretically prove the convergence of our DuMeta++, ensuring stability. Experiments on diverse datasets (iSeg-2019, IBIS, OASIS, ADNI) under few-shot settings demonstrate that DuMeta++ outperforms existing methods in cross-age generalization. Code will be available at https://github.com/ladderlab-xjtu/DuMeta++.

Problem

• brain tissue segmentation
• cross-age generalization
• few-shot learning
• longitudinal consistency
• MRI

Innovation

• dual meta-learning
• few-shot segmentation
• cross-age generalization
• memory-bank regularization
• age-agnostic representation