超声影像组学技术在评估胎肺成熟度中的应用

摘要/Abstract

摘要：

目的：基于超声影像组学技术,建立胎肺成熟度评估模型,并验证其预测新生儿发生呼吸系统疾病风险的价值。方法：本研究纳入295名孕妇（单胎）,每位孕妇收集1张分娩前72 h内采集的胎肺超声标准图像（四腔心切面）。依据收集胎肺超声图像当天的孕周,将295名孕妇图像分为组1（<36周）和组2（36~37周）。组1（66张）和组2图像（229张）再进一步分别分组为训练集（40、26张）和验证集（95、134张）。基于超声影像组学技术,对训练集胎肺超声图像进行胎肺纹理分析,结合妊娠并发症的有无,提取高通量影像组学特征,以新生儿预后结果为金标准,分别建立适用于不同孕周的胎肺成熟度评估模型,随后在相应的验证集中,验证胎肺成熟度模型预测新生儿发生呼吸系统疾病的风险。结果：本研究新生儿中,23.4%（69例）发生了呼吸系统疾病,6.1%（18例）发生了新生儿呼吸窘迫综合征。基于<36周和36~37孕周的胎肺成熟度模型,预测相应孕周的灵敏度分别为83.3%（组1）和75.0%（组2）,特异度为84.6%（组1）和78.3%（组2）;准确率为80.8%（组1）和77.2%（组2）。结论：基于超声影像组学技术胎肺成熟度评估模型具有一定的预测新生儿发生呼吸系统疾病的效能,这是无创性评估胎肺成熟度的一种新方法。

关键词:影像组学,胎肺成熟度,新生儿呼吸系统疾病,胎肺超声,妊娠并发症

Abstract:

Objective:To establish a model with ultrasound-based radiomics technology for evaluating fetal lung maturity and verify its efficacy in predicting risk of newborn respiratory disease.Methods:A total of 295 singleton pregnancies were enrolled, and fetal lung ultrasound images (four-chamber view) of each fetal were obtained within 72 hours before delivery. The 295 images were divided into 2 groups according to gestational age (GA) of the day fetal lung ultrasound images collected on examination day: Group 1(GA <36 weeks) and Group 2(GA 36-37 weeks). Images of Group 1 （66） and Group 2 （229） were further grouped into training set（40, 26） and validation set（95, 134）,respectively. High throughput radiomics features were extracted from each fetal lung ultrasound image by fetal lung texture analysis based on ultrasound-based radiomics technology. Based on outcomes of fetus, diagnostic models by Group 1 or 2 for predicting risk of newborn respiratory disease were established combined with pregnancy complications using training set of Group 1 or Group 2, respectively, and the predictive efficacy were verified in correspondingly validation set.Results:The diagnostic performance of models by Group1 and 2 were as follows: sensitivity were 83.3%(Group 1) and 75.0%, respectively; specificity were 84.6% and 78.3%,respectively; accuracy were 80.8% and 77.2%, respectively.Conclusions:The fetal lung maturity evaluation model based on ultrasonic-based radiomics technology is a new method for noninvasive evaluation of fetal lung maturity, Which have certain efficacy in predicting newborn respiratory disease.

Key words:Radiomics,Fetal lung maturity,Neonatal respiratory morbidity,Fetal lung ultrasound,Pregnancy complications

中图分类号:

R541.7

杜燕然, 焦景, 任芸芸, 周建桥. 超声影像组学技术在评估胎肺成熟度中的应用[J]. 诊断学理论与实践, 2022, 21(03): 326-330.

DU Yanran, JIAO Jing, REN Yunyun, ZHOU Jianqiao. Application of ultrasound-based radiomics technology in the evaluation of fetal lung maturity[J]. Journal of Diagnostics Concepts & Practice, 2022, 21(03): 326-330.

图/表5

图1

分组	截点	发生新生儿呼吸系统疾病	无新生儿呼吸系统疾病
组1
训练集	0.5	0.881（0.049~0.986）	0.159（0.033~0.621）
验证集		0.813（0.213~0.981）	0.346（0.078~0.630）
组2
训练集	0.5	0.538（0.290~0.748）	0.414（0.182~0.763）
验证集		0.543（0.324~0.733）	0.403（0.206~0.742）

表1

分组	发生新生儿呼吸系统疾病		无新生儿呼吸系统疾病
组1
训练集	20（83.3%）	4（16.7%）	1（6.3%）	15（93.7%）
验证集	10（76.9%）	3（23.1%）	2（15.4%）	11（84.6%）
组2
训练集	13（81.2%）	3（18.8%）	30（25.4%）	88（74.6%）
验证集	12（75.0%）	4（25.0%）	18（22.8%）	61（77.2%）

表2

图2

分组	新生儿呼吸系统疾病例数（n）	灵敏度（%）	特异度（%）	准确度（%）	阳性预测值（%）	阴性预测值（%）	阳性似然比	阴性似然比
组1（n=26）	13	83.3	84.6	80.8	76.9	84.6	5.4	0.2
组2（n=95）	16	75.0	78.3	77.2	40.0	93.8	3.5	0.3

表3

参考文献22

[1]	Teune MJ, Bakhuizen S, Gyamfi Bannerman C, et al. A systematic review of severe morbidity in infants born late preterm[J]. Am J Obstet Gynecol, 2011, 205(4):374,e1-e9.
[2]	Spong CY, Mercer BM, D′Alton M, et al. Timing of indicated late-preterm and early-term birth[J]. Obstet Gynecol, 2011, 118(2 Pt 1):323-333. doi:10.1097/AOG.0b013e3182255999 URL
[3]	Clark SL, Miller DD, Belfort MA, et al. Neonatal and maternal outcomes associated with elective term delivery[J]. Am J Obstet Gynecol, 2009, 200(2):156,e1-e4.
[4]	Sengupta S, Carrion V, Shelton J, et al. Adverse neonatal outcomes associated with early-term birth[J]. JAMA Pe-diatr, 2013, 167(11):1053-1059.
[5]	Serizawa M, Maeda K. Noninvasive fetal lung maturity prediction based on ultrasonic gray level histogram width[J]. Ultrasound Med Biol, 2010, 36(12):1998-2003. doi:10.1016/j.ultrasmedbio.2010.08.011pmid:20950934
[6]	Prakash KN, Ramakrishnan AG, Suresh S, et al. Fetal lung maturity analysis using ultrasound image features[J]. IEEE Trans Inf Technol Biomed, 2002, 6(1):38-45. doi:10.1109/4233.992160 URL
[7]	Cosmi EV, Anceschi MM, Cosmi E, et al. Ultrasonographic patterns of fetal breathing movements in normal pregnancy[J]. Int J Gynaecol Obstet, 2003, 80(3):285-290. doi:10.1016/S0020-7292(02)00384-3 URL
[8]	Maeda K, Utsu M, Yamamoto N, et al. Echogenicity of fetal lung and liver quantified by the grey-level histogram width[J]. Ultrasound Med Biol, 1999, 25(2):201-208. pmid:10320309
[9]	La Torre R, Cosmi E, Anceschi MH, et al. Preliminary report on a new and noninvasive method for the assessment of fetal lung maturity[J]. J Perinat Med, 2003, 31(5):431-434.
[10]	Insana MF, Garra BS, Rosenthal SJ, et al. Quantitative ultrasonography[J]. Med Prog Technol, 1989, 15(3-4):141-153. pmid:2533654
[11]	Sanz-Cortés M, Figueras F, Bonet-Carne E, et al. Fetal brain MRI texture analysis identifies different microstructural patterns in adequate and small for gestational age fetuses at term[J]. Fetal Diagn Ther, 2013, 33(2):122-129. doi:10.1159/000346566pmid:23391747
[12]	Du Y, Fang Z, Jiao J, et al. Application of ultrasound-based radiomics technology in fetal-lung-texture analysis in pregnancies complicated by gestational diabetes and/or pre-eclampsia[J]. Ultrasound Obstet Gynecol, 2021, 57(5):804-812. doi:10.1002/uog.22037 URL
[13]	Tekesin I, Anderer G, Hellmeyer L, et al. Assessment of fetal lung development by quantitative ultrasonic tissue characterization: a methodical study[J]. Prenat Diagn, 2004, 24(9):671-676. doi:10.1002/pd.951 URL
[14]	Palacio M, Bonet-Carne E, Cobo T, et al. Prediction of neonatal respiratory morbidity by quantitative ultrasound lung texture analysis: a multicenter study[J]. Am J Obstet Gynecol, 2017, 217(2):196,e1-e14.
[15]	Ahmed B, Konje JC. Fetal lung maturity assessment: A historic perspective and Non - invasive assessment using an automatic quantitative ultrasound analysis (a potentially useful clinical tool)[J]. Eur J Obstet Gynecol Reprod Biol, 2021, 258:343-347. doi:10.1016/j.ejogrb.2021.01.025 URL
[16]	Sengupta S, Carrion V, Shelton J, et al. Adverse neonatal outcomes associated with early-term birth[J]. JAMA Pe-diatr, 2013, 167(11):1053-1059.
[17]	Edwards MO, Kotecha SJ, Kotecha S. Respiratory distress of the term newborn infant[J]. Paediatr Respir Rev, 2013, 14(1):29-37. doi:10.1016/j.prrv.2012.02.002pmid:23347658
[18]	Tita ATN, Jablonski KA, Bailit JL, et al. Neonatal outcomes of elective early-term births after demonstrated fetal lung maturity[J]. Am J Obstet Gynecol, 2018, 219(3):296.
[19]	Spong CY, Mercer BM, D′Alton M, et al. Timing of indicated late-preterm and early-term birth[J]. Obstet Gynecol, 2011, 118(2 Pt 1):323-333. doi:10.1097/AOG.0b013e3182255999 URL
[20]	ACOG Committee Opinion No. 764 Summary: Medically Indicated Late-Preterm and Early-Term Deliveries[J]. Obstet Gynecol, 2019, 133(2):400-403. doi:10.1097/AOG.0000000000003084 URL
[21]	Towers CV, Freeman RK, Nageotte MP, et al. The case for amniocentesis for fetal lung maturity in late-preterm and early-term gestations[J]. Am J Obstet Gynecol, 2014, 210(2):95-96. doi:10.1016/j.ajog.2013.10.004pmid:24139938
[22]	Simon AE, Uddin SG. National trends in primary cesa-rean delivery, labor attempts, and labor success, 1990-2010[J]. Am J Obstet Gynecol,2013, 209(6):554,e1-e8.