Deep learning (DL) is increasingly used to solve ill-posed inverse problems in medical imaging, such as reconstruction from noisy and/or incomplete data, as DL offers advantages over conventional methods that rely on explicit image features and hand engineered priors. However, supervised DL-based methods may achieve poor performance when the test data deviates from the training data, for example, when it has pathologies not encountered in the training data. Furthermore, DL-based image reconstructions do not always incorporate the underlying forward physical model, which may improve performance. Therefore, in this work we introduce a novel approach, called fidelity imposed network edit (FINE), which modifies the weights of a pre-trained reconstruction network for each case in the testing dataset. This is achieved by minimizing an unsupervised fidelity loss function that is based on the forward physical model. FINE is applied to two important inverse problems in neuroimaging: quantitative susceptibility mapping (QSM) and under-sampled image reconstruction in MRI. Our experiments demonstrate that FINE can improve reconstruction accuracy.

中文翻译：

---

保真强加网络编辑 (FINE) 用于解决不适定图像重建


深度学习 (DL) 越来越多地用于解决医学成像中的不适定逆问题，例如从噪声和/或不完整数据中进行重建，因为深度学习比依赖显式图像特征和手工设计先验的传统方法具有优势。然而，当测试数据偏离训练数据时，例如，当它具有训练数据中未遇到的病症时，基于监督的深度学习方法可能会获得较差的性能。此外，基于深度学习的图像重建并不总是包含底层的前向物理模型，这可能会提高性能。因此，在这项工作中，我们引入了一种称为保真度强加网络编辑（FINE）的新颖方法，它修改测试数据集中每个案例的预训练重建网络的权重。这是通过最小化基于前向物理模型的无监督保真度损失函数来实现的。 FINE 应用于神经影像学中两个重要的逆问题：定量磁化率绘图 (QSM) 和 MRI 中的欠采样图像重建。我们的实验表明 FINE 可以提高重建精度。