Institute of Computer Graphics and Vision, Graz University of Technology, Inffeldgasse 16, Graz 8010, Austria; Computer Algorithms for Medicine Laboratory, Graz, Austria; BioTechMed-Graz, Mozartgasse 12/II, Graz 8010, Austria; Research Unit Experimental Neurotraumatology, Department of Neurosurgery, Medical University Graz, Auenbruggerplatz 2(2), Graz 8036, Austria.
Department of Neurosurgery, Medical University of Graz, Auenbruggerplatz 29, Graz, Austria.
Med Image Anal. 2021 Oct;73:102171. doi: 10.1016/j.media.2021.102171. Epub 2021 Jul 20.
A fast and fully automatic design of 3D printed patient-specific cranial implants is highly desired in cranioplasty - the process to restore a defect on the skull. We formulate skull defect restoration as a 3D volumetric shape completion task, where a partial skull volume is completed automatically. The difference between the completed skull and the partial skull is the restored defect; in other words, the implant that can be used in cranioplasty. To fulfill the task of volumetric shape completion, a fully data-driven approach is proposed. Supervised skull shape learning is performed on a database containing 167 high-resolution healthy skulls. In these skulls, synthetic defects are injected to create training and evaluation data pairs. We propose a patch-based training scheme tailored for dealing with high-resolution and spatially sparse data, which overcomes the disadvantages of conventional patch-based training methods in high-resolution volumetric shape completion tasks. In particular, the conventional patch-based training is applied to images of high resolution and proves to be effective in tasks such as segmentation. However, we demonstrate the limitations of conventional patch-based training for shape completion tasks, where the overall shape distribution of the target has to be learnt, since it cannot be captured efficiently by a sub-volume cropped from the target. Additionally, the standard dense implementation of a convolutional neural network tends to perform poorly on sparse data, such as the skull, which has a low voxel occupancy rate. Our proposed training scheme encourages a convolutional neural network to learn from the high-resolution and spatially sparse data. In our study, we show that our deep learning models, trained on healthy skulls with synthetic defects, can be transferred directly to craniotomy skulls with real defects of greater irregularity, and the results show promise for clinical use. Project page: https://github.com/Jianningli/MIA.
在颅骨修复术中，人们非常希望能够快速且全自动地设计出针对患者个体的 3D 打印颅骨植入物 - 这是一种用于修复颅骨缺损的过程。我们将颅骨缺损修复表述为一个 3D 体积形状完成任务，其中部分颅骨体积被自动完成。完成的颅骨与部分颅骨之间的差异即为需要修复的缺损；换句话说，就是可用于颅骨修复术的植入物。为了完成体积形状完成任务，我们提出了一种完全基于数据驱动的方法。在一个包含 167 个高分辨率健康颅骨的数据库上，进行了颅骨形状的监督学习。在这些颅骨中，合成的缺陷被注入以创建训练和评估数据对。我们提出了一种基于补丁的训练方案，专门用于处理高分辨率和空间稀疏的数据，克服了传统基于补丁的训练方法在高分辨率体积形状完成任务中的缺点。特别是，传统的基于补丁的训练方法适用于高分辨率图像，并且在分割等任务中证明是有效的。然而，我们展示了传统基于补丁的训练方法在形状完成任务中的局限性，因为目标的整体形状分布必须通过从目标中裁剪出的子体积来学习，而这无法被有效地捕捉。此外，卷积神经网络的标准密集实现往往在稀疏数据（如颅骨）上表现不佳，因为颅骨的体素占有率很低。我们提出的训练方案鼓励卷积神经网络从高分辨率和空间稀疏数据中学习。在我们的研究中，我们表明，我们在带有合成缺陷的健康颅骨上训练的深度学习模型可以直接转移到具有更大不规则性的真实缺陷的颅骨切开术中，并且结果显示出了临床应用的潜力。项目页面：https://github.com/Jianningli/MIA。
