Kaftan Paul, Heinrich Mattias P, Hansen Lasse, Rasche Volker, Kestler Hans A, Bigalke Alexander
Institute of Medical Systems Biology, Ulm University, Albert-Einstein-Allee 11, 89081, Ulm, Germany.
Institute of Medical Informatics, University of Lübeck, Ratzeburger Allee 160, 23562, Lübeck, Germany.
Int J Comput Assist Radiol Surg. 2025 Mar;20(3):465-473. doi: 10.1007/s11548-024-03310-z. Epub 2025 Jan 7.
Lung fissure segmentation on CT images often relies on 3D convolutional neural networks (CNNs). However, 3D-CNNs are inefficient for detecting thin structures like the fissures, which make up a tiny fraction of the entire image volume. We propose to make lung fissure segmentation more efficient by using geometric deep learning (GDL) on sparse point clouds.
We abstract image data with sparse keypoint (KP) clouds. We train GDL models to segment the point cloud, comparing three major paradigms of models (PointNets, graph convolutional networks (GCNs), and PointTransformers). From the sparse point segmentations, 3D meshes of the objects are reconstructed to obtain a dense surface. The state-of-the-art Poisson surface reconstruction (PSR) makes up most of the time in our pipeline. Therefore, we propose an efficient point cloud to mesh autoencoder (PC-AE) that deforms a template mesh to fit a point cloud in a single forward pass. Our pipeline is evaluated extensively and compared to the 3D-CNN gold standard nnU-Net on diverse clinical and pathological data.
GCNs yield the best trade-off between inference time and accuracy, being faster with only increased error over the nnU-Net. Our PC-AE also achieves a favorable trade-off, being faster at the error compared to the PSR.
We present a KP-based fissure segmentation pipeline that is more efficient than 3D-CNNs and can greatly speed up large-scale analyses. A novel PC-AE for efficient mesh reconstruction from sparse point clouds is introduced, showing promise not only for fissure segmentation. Source code is available on https://github.com/kaftanski/fissure-segmentation-IJCARS.
CT图像上的肺裂分割通常依赖于三维卷积神经网络（3D-CNN）。然而，3D-CNN在检测像肺裂这样的薄结构时效率低下，因为肺裂在整个图像体积中只占很小一部分。我们建议通过在稀疏点云上使用几何深度学习（GDL）来提高肺裂分割的效率。
我们用稀疏关键点（KP）云抽象图像数据。我们训练GDL模型来分割点云，比较三种主要的模型范式（点网、图卷积网络（GCN）和点变换器）。从稀疏点分割中，重建对象的三维网格以获得密集表面。我们流程中大部分时间由最先进的泊松表面重建（PSR）占据。因此，我们提出一种高效的点云到网格自动编码器（PC-AE），它在单次前向传播中使模板网格变形以拟合点云。我们的流程在多种临床和病理数据上进行了广泛评估，并与3D-CNN金标准nnU-Net进行了比较。
GCN在推理时间和准确性之间实现了最佳平衡，比nnU-Net更快，只是误差略有增加。我们的PC-AE也实现了良好的平衡，与PSR相比，在相同误差下速度更快。
我们提出了一种基于KP的肺裂分割流程，它比3D-CNN更高效，并且可以大大加快大规模分析的速度。还介绍了一种用于从稀疏点云进行高效网格重建的新型PC-AE，它不仅在肺裂分割方面显示出前景。源代码可在https://github.com/kaftanski/fissure-segmentation-IJCARS上获取。
