[论文] Anatomy-Informed Neural Networks: Encoding Anatomic Priors in Loss and...
论文概要
研究领域: CV 作者: David P. Stonko 发布时间: 2026-08-21 arXiv: 2608.21332
中文摘要
解剖学的深度学习模型在数值上可能合理但在解剖学上不可能,且在数据稀缺时泛化能力差。我们引入了解剖学知情神经网络(AINN),其中软解剖先验以惩罚项形式进入损失函数(例如,分支惩罚将肾移植动脉起源于髂动脉而非主动脉视为意外而非不可能),直接类比于物理知情神经网络;硬解剖先验(例如,血管的连续性)被构建进架构和状态表征中,使此类无效预测在先验允许架构强制的地方在构造上不可能。我们在一个数据有限的临床测试案例上开发它:当硬导丝经腔内引入时,主动脉-髂动脉树如何变形。这对当代主动脉手术很重要,也将影响自主血管内导航。我们将血管中心线和导丝路径从R³提升到Lie群SE(3)中的标架曲线,并将Cosserat杆导丝与通过单侧管腔接触不等式耦合的、由扭曲度调制的、解剖学锚定的血管耦合。预测是耦合弹性能量的约束最小化器,接触力作为其Lagrange乘子。监督是预测投影(通过C臂几何)与观测血管造影之间的Wasserstein-2最优传输损失,因此2D血管造影可以训练3D预测。运动学、损失和投影针对已知真值验证;力学求解器仅针对其自身最优性条件验证,预测位移尚未网格收敛。在此,没有训练网络。未来工作将将此in silico模型转移到真实CT扫描,并测试它是否提高预测准确性并减少所需训练数据。
原文摘要
Deep-learning models of anatomy can be numerically plausible yet anatomically impossible, and they generalize poorly when data are scarce. We introduce Anatomy-Informed Neural Networks (AINN), in which soft anatomic priors enter as penalty terms in the loss (e.g., a branching penalty that treats a renal transplant artery off the iliac instead of the aorta as unexpected rather than impossible), in direct analogy to a physics-informed neural network, and hard anatomic priors (e.g., continuity of the vessel) are built into the architecture and state representation, making such invalid predictions impossible by construction wherever the prior admits architectural enforcement. We develop it on a clinical test case with limited data: how the aortoiliac tree deforms when a stiff wire is introduce...
--- *自动采集于 2026-08-25*
#论文 #arXiv #CV #小凯