Chair for Computer Aided Medical Procedures & Augmented Reality, Munich, Germany.
Munich Data Science Institute, Munich, Germany.
Int J Comput Assist Radiol Surg. 2024 Jul;19(7):1339-1347. doi: 10.1007/s11548-024-03126-x. Epub 2024 May 15.
Ultrasound (US) imaging, while advantageous for its radiation-free nature, is challenging to interpret due to only partially visible organs and a lack of complete 3D information. While performing US-based diagnosis or investigation, medical professionals therefore create a mental map of the 3D anatomy. In this work, we aim to replicate this process and enhance the visual representation of anatomical structures.
We introduce a point cloud-based probabilistic deep learning (DL) method to complete occluded anatomical structures through 3D shape completion and choose US-based spine examinations as our application. To enable training, we generate synthetic 3D representations of partially occluded spinal views by mimicking US physics and accounting for inherent artifacts.
The proposed model performs consistently on synthetic and patient data, with mean and median differences of 2.02 and 0.03 in Chamfer Distance (CD), respectively. Our ablation study demonstrates the importance of US physics-based data generation, reflected in the large mean and median difference of 11.8 CD and 9.55 CD, respectively. Additionally, we demonstrate that anatomical landmarks, such as the spinous process (with reconstruction CD of 4.73) and the facet joints (mean distance to ground truth (GT) of 4.96 mm), are preserved in the 3D completion.
Our work establishes the feasibility of 3D shape completion for lumbar vertebrae, ensuring the preservation of level-wise characteristics and successful generalization from synthetic to real data. The incorporation of US physics contributes to more accurate patient data completions. Notably, our method preserves essential anatomical landmarks and reconstructs crucial injections sites at their correct locations.
超声（US）成像具有无辐射的优点，但由于部分器官可见度有限且缺乏完整的 3D 信息，因此其解读具有挑战性。在进行基于 US 的诊断或研究时，医疗专业人员会在脑海中创建 3D 解剖结构的心理图。在这项工作中，我们旨在复制这一过程并增强解剖结构的视觉表示。
我们引入了一种基于点云的概率深度学习（DL）方法，通过 3D 形状完成来完成被遮挡的解剖结构，并选择基于 US 的脊柱检查作为我们的应用。为了实现训练，我们通过模拟 US 物理并考虑固有伪影来生成部分遮挡的脊柱视图的合成 3D 表示。
该模型在合成数据和患者数据上表现一致，Chamfer 距离（CD）的平均值和中位数差异分别为 2.02 和 0.03。我们的消融研究表明基于 US 物理的数据生成的重要性，这反映在 CD 分别为 11.8 和 9.55 的平均值和中位数差异较大。此外，我们还证明了解剖学标志，如棘突（重建 CD 为 4.73）和关节突关节（与地面真实（GT）的平均距离为 4.96 毫米），在 3D 完成中得以保留。
我们的工作为腰椎 3D 形状完成确立了可行性，确保了按级别特征的保留以及从合成数据到真实数据的成功泛化。US 物理的纳入有助于更准确地完成患者数据。值得注意的是，我们的方法保留了重要的解剖学标志，并将关键的注射部位正确地重建到其位置。
