Medical Image Analysis ( IF 10.7 ) Pub Date : 2021-10-29 , DOI: 10.1016/j.media.2021.102292 R Han 1 , C K Jones 2 , J Lee 3 , P Wu 1 , P Vagdargi 4 , A Uneri 1 , P A Helm 5 , M Luciano 6 , W S Anderson 6 , J H Siewerdsen 7
|
Purpose
The accuracy of minimally invasive, intracranial neurosurgery can be challenged by deformation of brain tissue – e.g., up to 10 mm due to egress of cerebrospinal fluid during neuroendoscopic approach. We report an unsupervised, deep learning-based registration framework to resolve such deformations between preoperative MR and intraoperative CT with fast runtime for neurosurgical guidance.
Method
The framework incorporates subnetworks for MR and CT image synthesis with a dual-channel registration subnetwork (with synthesis uncertainty providing spatially varying weights on the dual-channel loss) to estimate a diffeomorphic deformation field from both the MR and CT channels. An end-to-end training is proposed that jointly optimizes both the synthesis and registration subnetworks. The proposed framework was investigated using three datasets: (1) paired MR/CT with simulated deformations; (2) paired MR/CT with real deformations; and (3) a neurosurgery dataset with real deformation. Two state-of-the-art methods (Symmetric Normalization and VoxelMorph) were implemented as a basis of comparison, and variations in the proposed dual-channel network were investigated, including single-channel registration, fusion without uncertainty weighting, and conventional sequential training of the synthesis and registration subnetworks.
Results
The proposed method achieved: (1) Dice coefficient = 0.82±0.07 and TRE = 1.2 ± 0.6 mm on paired MR/CT with simulated deformations; (2) Dice coefficient = 0.83 ± 0.07 and TRE = 1.4 ± 0.7 mm on paired MR/CT with real deformations; and (3) Dice = 0.79 ± 0.13 and TRE = 1.6 ± 1.0 mm on the neurosurgery dataset with real deformations. The dual-channel registration with uncertainty weighting demonstrated superior performance (e.g., TRE = 1.2 ± 0.6 mm) compared to single-channel registration (TRE = 1.6 ± 1.0 mm, p < 0.05 for CT channel and TRE = 1.3 ± 0.7 mm for MR channel) and dual-channel registration without uncertainty weighting (TRE = 1.4 ± 0.8 mm, p < 0.05). End-to-end training of the synthesis and registration subnetworks also improved performance compared to the conventional sequential training strategy (TRE = 1.3 ± 0.6 mm). Registration runtime with the proposed network was ∼3 s.
Conclusion
The deformable registration framework based on dual-channel MR/CT registration with spatially varying weights and end-to-end training achieved geometric accuracy and runtime that was superior to state-of-the-art baseline methods and various ablations of the proposed network. The accuracy and runtime of the method may be compatible with the requirements of high-precision neurosurgery.
中文翻译：
使用无监督双通道网络进行可变形 MR-CT 图像配准，用于神经外科指导
目的
微创颅内神经外科手术的准确性可能会受到脑组织变形的挑战，例如，在神经内窥镜手术过程中，由于脑脊液流出，脑组织变形最多可达 10 毫米。我们报告了一种无监督的、基于深度学习的配准框架，以解决术前 MR 和术中 CT 之间的这种变形，并具有快速的神经外科指导运行时间。
方法
该框架将 MR 和 CT 图像合成子网络与双通道配准子网络（合成不确定性为双通道损失提供空间变化的权重）结合起来，以估计来自 MR 和 CT 通道的微分同胚变形场。提出了一种端到端训练来联合优化合成子网络和配准子网络。使用三个数据集对所提出的框架进行了研究：（1）具有模拟变形的配对 MR/CT； (2) MR/CT与真实变形配对； (3)具有真实变形的神经外科数据集。采用两种最先进的方法（对称归一化和体素变形）作为比较的基础，并研究了所提出的双通道网络的变化，包括单通道配准、无不确定性加权的融合以及传统的顺序训练合成和注册子网络。
结果
该方法实现了：（1）在模拟变形的配对 MR/CT 上，Dice 系数 = 0.82±0.07，TRE = 1.2 ± 0.6 mm； (2) 在具有真实变形的配对 MR/CT 上，Dice 系数 = 0.83 ± 0.07，TRE = 1.4 ± 0.7 mm； (3) 在具有真实变形的神经外科数据集上，Dice = 0.79 ± 0.13 和 TRE = 1.6 ± 1.0 mm。与单通道配准（TRE = 1.6 ± 1.0 mm，CT 通道p < 0.05，TRE = 1.3 ± 0.7 mm）相比，具有不确定性加权的双通道配准表现出优越的性能（例如，TRE = 1.2 ± 0.6 mm）。 MR 通道）和无不确定性加权的双通道配准（TRE = 1.4 ± 0.8 mm， p < 0.05）。与传统的顺序训练策略（TRE = 1.3 ± 0.6 mm）相比，合成和配准子网络的端到端训练也提高了性能。建议网络的注册运行时间为 ∼3 秒。
结论
基于具有空间变化权重和端到端训练的双通道 MR/CT 配准的可变形配准框架实现了优于最先进的基线方法和所提出网络的各种消融的几何精度和运行时间。该方法的准确性和运行时间可以满足高精度神经外科的要求。
京公网安备 11010802027423号