深度学习在CT扫描中分割49块选定骨骼:基于PET/CT的骨转移瘤三维自动量化的第一步
本研究训练卷积神经网络对CT扫描中的49块骨骼进行自动分割,并在46例前列腺癌患者的CT上测试。自动与手动分割的Sørensen-Dice指数中位数为0.83–0.88,观察者内体积差异自动分割小于手动分割,作者认为该方法快速、自动且准确,似乎是开发临床可用骨转移瘤量化流程的有价值的第一步。
要点
- CNN自动分割49块骨骼,与手动分割的Sørensen-Dice指数中位数为0.83–0.88。
- 观察者内体积差异CNN分割小于手动分割:Th7为2%对14%,L3为7%对8%。
- 骶骨、第7肋、胸骨的观察者内体积差异CNN分别为1%、1%、3%,手动分别为3%、6%、5%。
核心结论
- 基于深度学习的CT骨骼自动分割方法可快速、自动地提供高度准确的骨体积,似乎是开发临床有用处理流程——以可靠骨骼分割作为骨转移瘤量化关键部分——的有价值的第一步。
研究设计与样本
- 卷积神经网络(CNN)使用100例CT扫描的手动分割训练分割49块骨骼,随后在46例前列腺癌患者中测试,这些患者间隔不到三周接受了18F-胆碱PET/CT和18F-NaF PET/CT;由一名经验丰富的医师对5块骨骼进行手动分割作为对比,采用Sørensen-Dice指数(SDI)评估空间重叠准确度,采用Bland-Altman法评估可重复性。
关键影像表现/征象
- 5块选定骨骼的CNN与手动分割中位(SD)体积分别为:Th7 41(3.8)对36(5.1),L3 76(13)对75(9.2),骶骨284(40)对283(26),第7肋33(3.9)对31(4.8),胸骨80(11)对72(9.2)。
- 中位SDI分别为Th7 0.86、L3 0.85、骶骨0.88、第7肋0.84、胸骨0.83。
诊断标准与关键数值
- 空间重叠准确度采用Sørensen-Dice指数(SDI)评估,中位SDI:Th7 0.86、L3 0.85、骶骨0.88、第7肋0.84、胸骨0.83。
- 可重复性采用Bland-Altman法评估;两次CNN分割间的平均体积差异(体积差/平均体积)脊柱和肋骨为5–6%,其他骨骼≤3%。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
PURPOSE: The aim of this study was to develop a deep learning-based method for segmentation of bones in CT scans and test its accuracy compared to manual delineation, as a first step in the creation of an automated PET/CT-based method for quantifying skeletal tumour burden.
METHODS: Convolutional neural networks (CNNs) were trained to segment 49 bones using manual segmentations from 100 CT scans. After training, the CNN-based segmentation method was tested on 46 patients with prostate cancer, who had undergone 18F-choline-PET/CT and 18F-NaF PET/CT less than three weeks apart. Bone volumes were calculated from the segmentations. The network's performance was compared with manual segmentations of five bones made by an experienced physician. Accuracy of the spatial overlap between automated CNN-based and manual segmentations of these five bones was assessed using the Sørensen-Dice index (SDI). Reproducibility was evaluated applying the Bland-Altman method.
RESULTS: The median (SD) volumes of the five selected bones were by CNN and manual segmentation: Th7 41 (3.8) and 36 (5.1), L3 76 (13) and 75 (9.2), sacrum 284 (40) and 283 (26), 7th rib 33 (3.9) and 31 (4.8), sternum 80 (11) and 72 (9.2), respectively. Median SDIs were 0.86 (Th7), 0.85 (L3), 0.88 (sacrum), 0.84 (7th rib) and 0.83 (sternum). The intraobserver volume difference was less with CNN-based than manual approach: Th7 2% and 14%, L3 7% and 8%, sacrum 1% and 3%, 7th rib 1% and 6%, sternum 3% and 5%, respectively. The average volume difference measured as ratio volume difference/mean volume between the two CNN-based segmentations was 5-6% for the vertebral column and ribs and ≤3% for other bones.
CONCLUSION: The new deep learning-based method for automated segmentation of bones in CT scans provided highly accurate bone volumes in a fast and automated way and, thus, appears to be a valuable first step in the development of a clinical useful processing procedure providing reliable skeletal segmentation as a key part of quantification of skeletal metastases.
引用
- Lindgren Belal S, Sadik M, Kaboteh R, et al. Deep learning for segmentation of 49 selected bones in CT scans: First step in automated PET/CT-based 3D quantification of skeletal metastases. Eur J Radiol. 2019. PMID: 30927965. doi:10.1016/j.ejrad.2019.01.028
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。