Skull Segmentation from CBCT Images via Voxel-Based Rendering.
Skull Segmentation from CBCT Images via Voxel-Based Rendering.
复制标题
基于体素绘制的CBCT图像颅骨分割。
DOI：
10.1007/978-3-030-87589-3_63
复制
发表时间：
2021-09
期刊：
影响因子：
--
通讯作者：
Xia JJ
中科院分区：
文献类型：
--
作者：
Liu Q;Lian C;Xiao D;Ma L;Deng H;Chen X;Shen D;Yap PT;Xia JJ
三维锥形束CT图像的颅骨分割对于颅颌面畸形的诊断和治疗具有重要意义。基于卷积神经网络（CNN）的方法目前在体积图像分割中占主导地位，但是这些方法受到有限的GPU存储器和大的图像尺寸（例如，512 × 512 × 448）。典型的临时策略（例如下采样或补丁裁剪）会由于对局部细节或全局上下文信息的捕获不足而降低分割准确性。其他方法如全局-局部网络（GLNet）专注于神经网络的改进，旨在以GPU内存高效的方式将局部细节和全局上下文信息联合收割机结合起来。然而，所有这些方法都是在规则网格上操作的，这对于体积图像分割来说是计算效率低下的。在这项工作中，我们提出了一种新的基于体素Rend的网络（VR-U-Net），通过将3D U-Net的内存效率变体与基于体素的渲染（VoxelRend）模块相结合，该模块通过基于体素的预测在非规则网格上细化局部细节。建立在相对粗糙的特征图上，VoxelRend模块以GPU内存消耗的一小部分实现了分割精度的显着提高。我们在从当地医院收集的高分辨率CBCT数据集上评估了我们提出的VR-U-Net颅骨分割任务。实验结果表明，VR-U-Net算法在节省内存的情况下获得了高质量的分割结果，具有较好的实用价值。
Skull segmentation from three-dimensional (3D) cone-beam computed tomography (CBCT) images is critical for the diagnosis and treatment planning of the patients with craniomaxillofacial (CMF) deformities. Convolutional neural network (CNN)-based methods are currently dominating volumetric image segmentation, but these methods suffer from the limited GPU memory and the large image size (e.g., 512 × 512 × 448). Typical ad-hoc strategies, such as down-sampling or patch cropping, will degrade segmentation accuracy due to insufficient capturing of local fine details or global contextual information. Other methods such as Global-Local Networks (GLNet) are focusing on the improvement of neural networks, aiming to combine the local details and the global contextual information in a GPU memory-efficient manner. However, all these methods are operating on regular grids, which are computationally inefficient for volumetric image segmentation. In this work, we propose a novel VoxelRend-based network (VR-U-Net) by combining a memory-efficient variant of 3D U-Net with a voxel-based rendering (VoxelRend) module that refines local details via voxel-based predictions on non-regular grids. Establishing on relatively coarse feature maps, the VoxelRend module achieves significant improvement of segmentation accuracy with a fraction of GPU memory consumption. We evaluate our proposed VR-U-Net in the skull segmentation task on a high-resolution CBCT dataset collected from local hospitals. Experimental results show that the proposed VR-U-Net yields high-quality segmentation results in a memory-efficient manner, highlighting the practical value of our method.
影响因子：
3.8
作者：
Wang，Li;Chen，Ken Chung;Shen，Dinggang
通讯作者：
Shen，Dinggang
使用计算机辅助手术模拟 (CASS) 协议规划双颌正颌手术的算法。第 1 部分：规划顺序。
DOI：
10.1016/j.ijom.2015.06.006
发表时间：
2015-12
影响因子：
2.4
作者：
Xia JJ;Gateno J;Teichgraeber JF;Yuan P;Chen KC;Li J;Zhang X;Tang Z;Alfi DM
通讯作者：
Alfi DM