用神经场压缩心血管影像与血流数据，精度高且节省内存。
On the accuracy of implicit neural representations for cardiovascular anatomies and hemodynamic fields
- 用连续函数代替体素/网格，实现分辨率无关的高效表示。
- 血流压力误差≤1 mmHg，速度误差5-10 cm/s，解剖结构偏差<0.5 mm。
- SIREN等模型表现最佳，适合医学影像与血流模拟研究者。
隐式神经表示（INRs）作为一种新兴的知识表征、合成与压缩框架，通过将场信息编码为深度神经网络权重与偏置中的连续函数，取代传统的体素或网格结构表示，具备分辨率独立和高内存效率的优势。然而其在特定领域的准确性仍不明确。本文评估了当前先进INRs在压缩胸主动脉数值模拟所得的时空变化血流场及基于符号距离函数的心血管解剖表示中的性能。针对频谱偏差问题，探索了专用激活函数、固定与可训练位置编码以及非线性核的线性组合等策略。在真实胸主动脉的时空血流场中，INRs实现了约230倍的压缩比，压力最大绝对误差为1 mmHg，速度为5-10 cm/s，无需大量超参数调优。在48例胸主动脉解剖结构上，平均与最大绝对解剖偏差分别低于0.5 mm和1.6 mm。总体而言，SIREN、MFN-Gabor与MHE架构表现最优。代码与数据已开源：https://github.com/desResLab/nrf。
原文摘要 · Abstract (English)
Implicit neural representations (INRs, also known as neural fields) have recently emerged as a powerful framework for knowledge representation, synthesis, and compression. By encoding fields as continuous functions within the weights and biases of deep neural networks-rather than relying on voxel- or mesh-based structured or unstructured representations-INRs offer both resolution independence and high memory efficiency. However, their accuracy in domain-specific applications remains insufficiently understood. In this work, we assess the performance of state-of-the-art INRs for compressing hemodynamic fields derived from numerical simulations and for representing cardiovascular anatomies via signed distance functions. We investigate several strategies to mitigate spectral bias, including specialized activation functions, both fixed and trainable positional encoding, and linear combinations of nonlinear kernels. On realistic, space- and time-varying hemodynamic fields in the thoracic aorta, INRs achieved remarkable compression ratios of up to approximately 230, with maximum absolute errors of 1 mmHg for pressure and 5-10 cm/s for velocity, without extensive hyperparameter tuning. Across 48 thoracic aortic anatomies, the average and maximum absolute anatomical discrepancies were below 0.5 mm and 1.6 mm, respectively. Overall, the SIREN, MFN-Gabor, and MHE architectures demonstrated the best performance. Source code and data is available at https://github.com/desResLab/nrf.
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。