文献速递妇产影像US

三维超声全自动胎盘体积定量预测小于胎龄儿

Fully Automated Placental Volume Quantification From 3D Ultrasound for Prediction of Small-for-Gestational-Age Infants.
PubMed 文献 · J Ultrasound Med · 2021-09-23

研究基于442例单胎妊娠11-14周三维超声图像,构建卷积神经网络全自动分割胎盘并计算体积(PVCNN),用于预测小于胎龄儿(SGA10/SGA5),并与半自动VOCAL法(PVVOCAL)比较。PVCNN联合临床模型对SGA10和SGA5的预测AUC均有统计学显著提升,而PVVOCAL仅对SGA10未达显著。作者认为该全自动工具可将胎盘体积生物测量纳入床旁临床评估的多变量预测模型。

要点

  • CNN分割 pipeline 在独立测试集平均Dice评分0.88。
  • PVCNN预测SGA10的AUC为0.780,PVVOCAL为0.768。
  • PVCNN加入临床模型使SGA10的AUC从0.725显著提升(P < .05),PVVOCAL未达显著(P = .105)。

核心结论

  • 早孕期基于CNN分割 pipeline 测量的胎盘体积与未来小于胎龄儿显著相关,该全自动工具有望将胎盘体积生物测量纳入床旁临床评估的多变量预测模型,用于风险分层和患者咨询。

研究设计与样本

  • 研究纳入442例单胎妊娠,于11-14周妊娠行三维超声检查;CNN pipeline 在99/25幅图像上训练和测试,结合两个2D模型和一个3D模型及下采样/上采样架构,自动分割胎盘并计算PVCNN;以半自动VOCAL(GE-Healthcare)法所得PVVOCAL作为对比;采用多变量逻辑回归分类器预测SGA10/SGA5,并在测试集比较预测性能。

诊断标准与关键数值

  • CNN分割 pipeline 在独立测试集平均Dice评分为0.88;442例中SGA10为37例(8.4%),SGA5为18例(4.1%);调整模型预测SGA10的AUC:PVCNN=0.780,PVVOCAL=0.768;不含任何胎盘体积的临床模型AUC=0.725,加入PVCNN后AUC提升有统计学显著意义(P < .05),加入PVVOCAL后未达统计学显著(P = .105);预测SGA5时,PVCNN模型AUC=0.897,临床模型AUC=0.839(P = .015),PVVOCAL模型AUC=0.870(P = .039)。

临床应用与阅片要点

  • 作者认为该全自动工具可将胎盘体积生物测量纳入床旁临床评估,作为多变量预测模型的一部分用于风险分层和患者咨询。

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

OBJECTIVES: Early placental volume (PV) has been associated with small-for-gestational-age infants born under the 10th/5th centiles (SGA10/SGA5). Manual or semiautomated PV quantification from 3D ultrasound (3DUS) is time intensive, limiting its incorporation into clinical care. We devised a novel convolutional neural network (CNN) pipeline for fully automated placenta segmentation from 3DUS images, exploring the association between the calculated PV and SGA.

METHODS: Volumes of 3DUS obtained from singleton pregnancies at 11-14 weeks' gestation were automatically segmented by our CNN pipeline trained and tested on 99/25 images, combining two 2D and one 3D models with downsampling/upsampling architecture. The PVs derived from the automated segmentations (PVCNN ) were used to train multivariable logistic-regression classifiers for SGA10/SGA5. The test performance for predicting SGA was compared to PVs obtained via the semiautomated VOCAL (GE-Healthcare) method (PVVOCAL ).

RESULTS: We included 442 subjects with 37 (8.4%) and 18 (4.1%) SGA10/SGA5 infants, respectively. Our segmentation pipeline achieved a mean Dice score of 0.88 on an independent test-set. Adjusted models including PVCNN or PVVOCAL were similarly predictive of SGA10 (area under curve [AUC]: PVCNN = 0.780, PVVOCAL = 0.768). The addition of PVCNN to a clinical model without any PV included (AUC = 0.725) yielded statistically significant improvement in AUC (P < .05); whereas PVVOCAL did not (P = .105). Moreover, when predicting SGA5, including the PVCNN (0.897) brought statistically significant improvement over both the clinical model (0.839, P = .015) and the PVVOCAL model (0.870, P = .039).

CONCLUSIONS: First trimester PV measurements derived from our CNN segmentation pipeline are significantly associated with future SGA. This fully automated tool enables the incorporation of including placental volumetric biometry into the bedside clinical evaluation as part of a multivariable prediction model for risk stratification and patient counseling.

来源:J Ultrasound Med,查看原文。版权归原作者所有。

引用

  1. Schwartz N, Oguz I, Wang J, et al. Fully Automated Placental Volume Quantification From 3D Ultrasound for Prediction of Small-for-Gestational-Age Infants. J Ultrasound Med. 2021. PMID: 34553780. doi:10.1002/jum.15835

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。