Unsupervised supervoxel-based lung tumor segmentation across patient scans in hybrid PET/MRI

作者：

Highlights：

• Unsupervised framework for lung tumor segmentation in hybrid PET/MRI.

• Superpixel-based approach to utilize information across patients.

• Evaluation of common clustering approaches within the framework.

• Segmentation performance increases with the across-patients clustering.

摘要

•Unsupervised framework for lung tumor segmentation in hybrid PET/MRI.•Superpixel-based approach to utilize information across patients.•Evaluation of common clustering approaches within the framework.•Segmentation performance increases with the across-patients clustering.

论文关键词：Clustering,Unsupervised learning,Medical image segmentation,Tumor segmentation,Hybrid PET/MRI

论文评审过程：Received 10 September 2020, Revised 2 November 2020, Accepted 4 November 2020, Available online 29 November 2020, Version of Record 9 December 2020.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114244