Division of Nuclear Medicine and Molecular Imaging, Geneva University Hospital, Geneva, Switzerland.
Department of Integrative Oncology, BC Cancer Research Institute, Vancouver, British Columbia, Canada.
Med Phys. 2024 Jan;51(1):319-333. doi: 10.1002/mp.16615. Epub 2023 Jul 20.
PET/CT images combining anatomic and metabolic data provide complementary information that can improve clinical task performance. PET image segmentation algorithms exploiting the multi-modal information available are still lacking.
Our study aimed to assess the performance of PET and CT image fusion for gross tumor volume (GTV) segmentations of head and neck cancers (HNCs) utilizing conventional, deep learning (DL), and output-level voting-based fusions.
The current study is based on a total of 328 histologically confirmed HNCs from six different centers. The images were automatically cropped to a 200 × 200 head and neck region box, and CT and PET images were normalized for further processing. Eighteen conventional image-level fusions were implemented. In addition, a modified U2-Net architecture as DL fusion model baseline was used. Three different input, layer, and decision-level information fusions were used. Simultaneous truth and performance level estimation (STAPLE) and majority voting to merge different segmentation outputs (from PET and image-level and network-level fusions), that is, output-level information fusion (voting-based fusions) were employed. Different networks were trained in a 2D manner with a batch size of 64. Twenty percent of the dataset with stratification concerning the centers (20% in each center) were used for final result reporting. Different standard segmentation metrics and conventional PET metrics, such as SUV, were calculated.
In single modalities, PET had a reasonable performance with a Dice score of 0.77 ± 0.09, while CT did not perform acceptably and reached a Dice score of only 0.38 ± 0.22. Conventional fusion algorithms obtained a Dice score range of [0.76-0.81] with guided-filter-based context enhancement (GFCE) at the low-end, and anisotropic diffusion and Karhunen-Loeve transform fusion (ADF), multi-resolution singular value decomposition (MSVD), and multi-level image decomposition based on latent low-rank representation (MDLatLRR) at the high-end. All DL fusion models achieved Dice scores of 0.80. Output-level voting-based models outperformed all other models, achieving superior results with a Dice score of 0.84 for Majority_ImgFus, Majority_All, and Majority_Fast. A mean error of almost zero was achieved for all fusions using SUV , SUV and SUV .
PET/CT information fusion adds significant value to segmentation tasks, considerably outperforming PET-only and CT-only methods. In addition, both conventional image-level and DL fusions achieve competitive results. Meanwhile, output-level voting-based fusion using majority voting of several algorithms results in statistically significant improvements in the segmentation of HNC.
PET/CT 图像结合解剖和代谢数据提供互补信息，可提高临床任务的性能。利用可用的多模态信息的 PET 图像分割算法仍然缺乏。
本研究旨在评估利用常规、深度学习（DL）和输出级投票融合对头部和颈部癌症（HNC）的大体肿瘤体积（GTV）进行 PET 和 CT 图像融合的性能。
本研究基于来自六个不同中心的 328 例经组织学证实的 HNC。图像被自动裁剪为 200×200 的头颈部区域框，对 CT 和 PET 图像进行归一化以进行进一步处理。实施了 18 种常规图像级融合。此外，还使用了修改后的 U2-Net 架构作为 DL 融合模型基线。使用三种不同的输入、层和决策级信息融合。采用同时真实和性能水平估计（STAPLE）和多数投票来合并不同的分割输出（来自 PET 和图像级和网络级融合），即输出级信息融合（基于投票的融合）。不同的网络以批处理大小为 64 的 2D 方式进行训练。使用分层（每个中心 20%）的数据集的 20%用于最终结果报告。计算了不同的标准分割指标和常规 PET 指标，如 SUV。
在单模态中，PET 表现出合理的性能，Dice 评分达到 0.77±0.09，而 CT 表现不佳，仅达到 0.38±0.22 的 Dice 评分。常规融合算法的 Dice 评分范围为[0.76-0.81]，其中基于导向滤波器的上下文增强（GFCE）在低端，各向异性扩散和 Karhunen-Loeve 变换融合（ADF）、多分辨率奇异值分解（MSVD）和基于潜在低秩表示的多层图像分解（MDLatLRR）在高端。所有的 DL 融合模型都达到了 0.80 的 Dice 评分。输出级投票的基于模型的模型表现优于所有其他模型，使用 Majority_ImgFus、 Majority_All 和 Majority_Fast 达到了 0.84 的 Dice 评分。使用 SUV、SUV 和 SUV ，所有融合都实现了几乎为零的平均误差。
PET/CT 信息融合为分割任务增加了显著价值，明显优于 PET 仅和 CT 仅方法。此外，常规图像级和 DL 融合都取得了有竞争力的结果。同时，使用几种算法的多数投票的输出级投票融合导致 HNC 分割的统计显著改善。
