分离共享与个体特征，提升医学图像重建精度与效率
Disentangled Learning Improves Implicit Neural Representations for Medical Reconstruction
- 将共享与个体特征解耦，仅优化个体编码器
- 在有限原始数据上预训练共享模块，无需高质量图像
- 测试时仅微调个体部分，避免遗忘，适合医疗影像应用
隐式神经表示（INRs）通过物理引导的无监督学习成为医学成像的强大范式。传统INRs为每个受试者从头训练整个网络，导致训练低效且成像质量不佳。基于初始化的方法虽尝试注入群体先验，但依赖高质量图像，且微调时易发生灾难性遗忘。本文提出DisINR，一种新型INR框架，显式分离共享与受试者特异性表示。DisINR引入共享编码器-解码器对和受试者特异性编码器，其特征联合解码以实现图像重建。通过集成可微分前向模型，它直接从少量原始测量数据预训练共享模块，无需预先获取高质量图像。测试时仅优化受试者特异性编码器，共享对保持冻结，有效保留学习到的先验。在三个代表性医学成像任务上的广泛评估表明，DisINR在重建精度和效率方面显著优于现有最先进INRs。
原文摘要 · Abstract (English)
Implicit neural representations (INRs) have emerged as a powerful paradigm for medical imaging via physics-informed unsupervised learning. Classical INRs optimize an entire network from scratch for each subject, leading to inefficient training and suboptimal imaging quality. Recent initialization-based approaches attempt to inject population priors into pre-trained networks, yet they rely on high-quality images and often suffer from catastrophic forgetting during fine-tuning. We present DisINR, a novel INR framework that explicitly disentangles shared and subject-specific representations. DisINR introduces a shared encoder-decoder pair and subject-specific encoders, whose features are jointly decoded for image reconstruction. By integrating differentiable forward models, it pre-trains the shared modules directly from limited raw measurements, removing the need for pre-acquired high-quality images. During test-time adaptation, only the subject-specific encoder is optimized, while the shared pair remains frozen, effectively preserving learned priors. Extensive evaluations on three representative medical imaging tasks show that DisINR significantly outperforms state-of-the-art INRs in both reconstruction accuracy and efficiency.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。