深度学习全卷积网络自动检测68Ga-DOTATATE PET/CT肝脏病灶
本研究回顾性纳入125例68Ga-DOTATATE PET/CT检查,其中57例有肝脏病灶、68例无病灶,采用2D U-Net卷积神经网络自动检测肝脏病灶。结果显示,像素滤波阈值20时平均阳性预测值最高(0.94±0.01),阈值5时平均灵敏度最高(0.74±0.02),阈值20时平均F1分数最高(0.79±0.01),阈值15时平均PR-AUC最高(0.73±0.03)。结论认为深度神经网络可自动检测68Ga-DOTATATE PET肝脏病灶,未来需改进标注方法、扩大样本量和训练方法以进一步提升性能。
要点
- 2D U-Net可在68Ga-DOTATATE PET/CT上自动检测肝脏病灶。
- 像素滤波阈值20时平均PPV最高(0.94±0.01),阈值5时平均灵敏度最高(0.74±0.02)。
- 阈值20时平均F1分数最高(0.79±0.01),阈值15时平均PR-AUC最高(0.73±0.03)。
核心结论
- 深度神经网络能够自动检测68Ga-DOTATATE PET中的肝脏病灶。
- 预期通过改进数据标注方法、增加样本量和训练方法可进一步提升检测性能。
研究设计与样本
- 回顾性研究纳入125例68Ga-DOTATATE PET/CT患者检查,其中57例有肝脏病灶、68例无病灶。
- 数据集随机分为训练集75例(36例异常、39例正常)、验证集25例(11例异常、14例正常)和测试集25例(11例异常、14例正常)。
- 肝脏病灶由医师采用改良PERCIST阈值标注,并通过梯度边缘检测定义边界。
- 2D U-Net独立训练5次、每次100,000次迭代,使用二元交叉熵与dice损失的线性组合及随机梯度下降算法。
- 采用5种不同像素面积阈值过滤噪声预测,评估PPV、灵敏度、F1分数和PR-AUC。
主要发现
- 共标注233个病灶,每例异常检查平均含4±2.75个病灶。
- 像素滤波阈值20产生最高平均PPV 0.94±0.01。
- 像素滤波阈值5产生最高平均灵敏度0.74±0.02。
- 像素滤波阈值20产生最高平均F1分数0.79±0.01。
- 像素滤波阈值15产生最高平均PR-AUC 0.73±0.03。
关键数值与推荐
- 肝脏病灶标注采用改良PERCIST阈值,边界通过梯度边缘检测定义。
- 像素面积阈值20时平均PPV最高(0.94±0.01),阈值5时平均灵敏度最高(0.74±0.02)。
- 像素面积阈值20时平均F1分数最高(0.79±0.01),阈值15时平均PR-AUC最高(0.73±0.03)。
临床怎么用
- 该自动化方法有望辅助识别68Ga-DOTATATE PET/CT上的肝脏转移灶,但需注意不同像素滤波阈值下PPV与灵敏度的权衡。
局限性
- 样本量有限,标注方法和训练方法仍需改进以进一步提升检测性能。
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)。
原文
BACKGROUND: Gastroenteropancreatic neuroendocrine tumors most commonly metastasize to the liver; however, high normal background 68Ga-DOTATATE activity and high image noise make metastatic lesions difficult to detect. The purpose of this study is to develop a rapid, automated and highly specific method to identify 68Ga-DOTATATE PET/CT hepatic lesions using a 2D U-Net convolutional neural network. METHODS: A retrospective study of 68Ga-DOTATATE PET/CT patient studies (n = 125; 57 with 68Ga-DOTATATE hepatic lesions and 68 without) was evaluated. The dataset was randomly divided into 75 studies for the training set (36 abnormal, 39 normal), 25 for the validation set (11 abnormal, 14 normal) and 25 for the testing set (11 abnormal, 14 normal). Hepatic lesions were physician annotated using a modified PERCIST threshold, and boundary definition by gradient edge detection. The 2D U-Net was trained independently five times for 100,000 iterations using a linear combination of binary cross-entropy and dice losses with a stochastic gradient descent algorithm. Performance metrics included: positive predictive value (PPV), sensitivity, F1 score and area under the precision-recall curve (PR-AUC). Five different pixel area thresholds were used to filter noisy predictions. RESULTS: A total of 233 lesions were annotated with each abnormal study containing a mean of 4 ± 2.75 lesions. A pixel filter of 20 produced the highest mean PPV 0.94 ± 0.01. A pixel filter of 5 produced the highest mean sensitivity 0.74 ± 0.02. The highest mean F1 score 0.79 ± 0.01 was produced with a 20 pixel filter. The highest mean PR-AUC 0.73 ± 0.03 was produced with a 15 pixel filter. CONCLUSION: Deep neural networks can automatically detect hepatic lesions in 68Ga-DOTATATE PET. Ongoing improvements in data annotation methods, increasing sample sizes and training methods are anticipated to further improve detection performance.
引用
- Wehrend J, Silosky M, Xing F, et al. Automated liver lesion detection in 68Ga DOTATATE PET/CT using a deep fully convolutional neural network. EJNMMI Res. 2021. PMID: 34601660. doi:10.1186/s13550-021-00839-x
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。