唤醒，然后缩放：用于MRI重建的微调适配
Awaken, Then Scale: Tiny Adaptation for MRI Reconstruction
浏览论文内容
中文总结 AI 辅助
本文提出休眠唤醒（DA）方法，通过拟合少量零权重特化MRI重建模型，在10%适配预算下比较动态封顶与固定减半等策略，显著提升PSNR并控制误差变化。
中文摘要 AI 辅助
休眠唤醒（Dormant Awakening, DA）通过将一小部分零权重拟合到一个标记切片来特化磁共振成像（MRI）重建模型。在十四个U形卷积网络（U-Net）和视觉变换器（ViT）源中，拟合1,004或10,560个权重分别带来平均峰值信噪比（PSNR）提升0.430和0.509 dB。我们分析了缩放拟合输出校正如何改变评估PSNR，然后使用校准校正大小来限制新图像上的变化。该限制在无评估参考的情况下，相对于源限制了均方根误差（RMSE）的变化。在10%适配预算下，我们比较了固定减半和动态封顶的DA、无限制稀疏适配和低秩适配（LoRA）。对照组在所有六个测试的架构-适配器设置中均增加了测量的校准-评估相关性。实验涉及一个构造的fastMRI到M4Raw迁移，使用固定源检查点和参与者面板。
英文摘要
Dormant Awakening (DA) specializes a magnetic resonance imaging (MRI) reconstruction model by fitting a small set of zero weights to one labeled slice. Across fourteen U-shaped convolutional network (U-Net) and vision transformer (ViT) sources, fitting 1,004 or 10,560 weights gives mean peak signal-to-noise ratio (PSNR) gains of .430 and .509 dB. We analyze how scaling the fitted output correction changes evaluation PSNR, then use the calibration correction size to cap changes on new images. The cap bounds changes in root mean squared error (RMSE) relative to the source without an evaluation reference. At 10\% adaptation budget, we compare fixed halving and dynamic capping for DA, unrestricted sparse adaptation and low-rank adaptation (LoRA). The controls increase both measured calibration-evaluation correlations across all six tested architecture-adapter settings. Experiments concern one constructed fastMRI-to-M4Raw shift with fixed source checkpoints and participant panels.
发表机构
- Technion-Israel Institute of Technology（以色列理工学院）
机构由 AI 辅助整理，请以论文原文为准。
