用于骨闪烁成像骨闪烁指数（BSI）计算的骨骼分割

Skeleton Segmentation on Bone Scintigraphy for BSI Computation.

作者信息

Yu Po-Nien, Lai Yung-Chi, Chen Yi-You, Cheng Da-Chuan

机构信息

Department of Biomedical Imaging and Radiological Science, China Medical University, Taichung 404, Taiwan.

Department of Nuclear Medicine, Feng Yuan Hospital Ministry of Health and Welfare, Taichung 420, Taiwan.

出版信息

Diagnostics (Basel). 2023 Jul 6;13(13):2302. doi: 10.3390/diagnostics13132302.

Bone Scan Index (BSI) is an image biomarker for quantifying bone metastasis of cancers. To compute BSI, not only the hotspots (metastasis) but also the bones have to be segmented. Most related research focus on binary classification in bone scintigraphy: having metastasis or none. Rare studies focus on pixel-wise segmentation. This study compares three advanced convolutional neural network (CNN) based models to explore bone segmentation on a dataset in-house. The best model is Mask R-CNN, which reaches the precision, sensitivity, and F1-score: 0.93, 0.87, 0.90 for prostate cancer patients and 0.92, 0.86, and 0.88 for breast cancer patients, respectively. The results are the average of 10-fold cross-validation, which reveals the reliability of clinical use on bone segmentation.

骨扫描指数（BSI）是一种用于量化癌症骨转移的图像生物标志物。为了计算BSI，不仅要对热点区域（转移灶）进行分割，还必须对骨骼进行分割。大多数相关研究集中于骨闪烁显像中的二分类：有转移或无转移。很少有研究关注逐像素分割。本研究比较了三种基于先进卷积神经网络（CNN）的模型，以在内部数据集上探索骨骼分割。最佳模型是Mask R-CNN，对于前列腺癌患者，其精度、灵敏度和F1分数分别达到0.93、0.87、0.90；对于乳腺癌患者，分别为0.92、0.86和0.88。结果是10折交叉验证的平均值，揭示了其在骨骼分割临床应用中的可靠性。