Intelligent prediction of neonatal birth weight based on prenatal sequential ultrasound dataYang

doi:10.3877/cma.j.issn.1672-6448.2025.08.006

Abstract

Abstract:

Objective

To establish a neonatal birth weight prediction model based on the XGBoost algorithm to provide a precise intelligent tool for clinical perinatal decision-making.

Methods

A retrospective analysis was conducted on data from 1018 singleton pregnancies who underwent routine prenatal examinations and delivered at the Yunnan Maternal and Child Health Hospital between January 2021 and April 2023. The dataset included maternal basic information, prenatal ultrasound examination parameters (53 variables in total) during the first, second, and third trimesters, as well as within 7 days before delivery, and neonatal data (gestational age at delivery and birth weight). An XGBoost prediction model was constructed, and its performance was evaluated using multidimensional metrics (mean percentage error, standard deviation of percentage error, mean absolute percentage error, root mean square error, mean absolute error, and clinical prediction accuracy). The XGBoost model was compared with 15 traditional prediction formulas and machine learning models (LightGBM, Logistic, and CatBoost). Feature importance analysis was also performed.

Results

The XGBoost model demonstrated strong performance in predicting neonatal birth weight, with a percentage error standard deviation of 8.21%, mean absolute percentage error of 6.34%, root mean square error of 245.14 g, and mean absolute error of 194.39 g, all error metrics were lower than those of other prediction methods. The clinical accuracy rate reached 71.54%, surpassing those of other predictive approaches. Compared to other benchmark models, the XGBoost model exhibited relatively balanced performance with an overestimation rate of 13.73% and an underestimation rate of 14.71%, both lower than those of Logistic (23.53%/28.43%), LightGBM (25.98%/24.02%), and CatBoost (17.16%/17.65%). Feature importance analysis identified 21 key predictors, among which fetal abdominal circumference before delivery, gestational age at delivery, and maximum amniotic fluid depth before delivery emerged as the most significant predictive variables (feature importance scores of 35, 22, and 8, respectively).

Conclusion

An XGBoost-based temporally correlated neonatal weight prediction model with robust stability and precision has been successfully constructed, and it outperforms traditional regression-based fetal weight prediction methods.

Key words: Prenatal ultrasound, Weight prediction, Artificial intelligence, XGBoost model

xian Li, Zhixi Huang, Bocheng Liang, Shuyuan Ouyang, Minglang Chen, Yingli Zhao, Weibo Ma, Jing Miao, Lei Wang, Ying Yuan. Intelligent prediction of neonatal birth weight based on prenatal sequential ultrasound dataYang[J]. Chinese Journal of Medical Ultrasound (Electronic Edition), 2025, 22(08): 721-732.

/ / Recommend

Add to citation manager EndNote|Ris|BibTeX

URL: https://chaosheng.cma-cmc.com.cn/EN/10.3877/cma.j.issn.1672-6448.2025.08.006

https://chaosheng.cma-cmc.com.cn/EN/Y2025/V22/I08/721

Figures/Tables 21

产前超声检查时间	纳入的超声检查数据参数
孕早期（11~13⁺⁶周）	GA，CRL，NT，FHR，MVP
孕中期（20~24⁺⁶周）	GA，BPD，HC，AC，FL，HL，CER，MVP，UA（Vmax、Vmin、S/D、RI、PI），MCA（Vmax、Vmin、S/D、PI、RI），FHR
孕晚期（28~34周）	GA，BPD，HC，AC，FL，MVP，UA（Vmax、Vmin、S/D、RI、PI），MCA（Vmax、Vmin、S/D、PI、RI），FHR
分娩前7 d内	GA，BPD，HC，AC，FL，MVP，UA（Vmax、Vmin、S/D、RI、PI），FHR

序号	作者	计算公式
1	Hadlock I^[14]	$lg (W) = 1.3596 + 0.0064 × H C + 0.0424 × A C + 0.174 × F L + 0.00061 × B P D × A C − 0.00386 × A C × F L$
2	Hadlock Ⅳ^[14]	$lg (W) = 1.326 − 0.00326 × A C × F L + 0.0107 × H L + 0.0438 × A C + 0.158 × F L$
3	Hadlock Ⅱ^[14]	$lg (W) = 1.304 + 0.05281 × A C + 0.1938 × F L − 0.004 × A C × F L$
4	Hadlock Ⅲ^[14]	$lg (W) = 1.335 − 0.0034 × A C × F L + 0.0316 × B P D + 0.0457 × A C + 0.1623 × F L$
5	Ott^[14]	$lg (W) = 0.04355 × H C + 0.05394 × A C − 0.0008582 × H C × A C + 1.2504 × F L A C + 0.9339$
6	Combs^[14]	$lg (W) = 0.23718 × A C 2 × F L + 0.03312 × H C 2$
7	Bernstein^[15]	$lg (W) = 1.42482 + 0.01165 × H C + 0.03949 × A C − 0.00028 × A C 2 + 0.14147 × F L − 0.00662 × F L 2$
8	Hadlock V^[14]	$lg (W) = 1.5662 − 0.0108 H C + 0.0468 × A C + 0.171 × F L + 0.00034 × H C 2 − 0.003685 × A C × F L$
9	Jordaan^[16]	$lg (W) = 2.3231 + 0.02904 × A C + 0.0079 × H C − 0.0058 × B P D$
10	ShepardⅡ^[14]	$lg (W) = 1.2508 + 0.166 × B P D + 0.046 × A C − 0.002646 × A C × B P D$
11	Johnson^[15]	$lg (W) = 1.21633 + 0.06076 × H C − 0.00075 × A C 2 + 0.02107 × A C + 0.05261 × F L$
12	Merz^[14]	$W = − 3200.40479 + 57.07186 × A C + 15.90391 × B P D 2$
13	Warsof^[14]	$lg (W) = 1.467 + 0.144 × B P D + 0.032 × A C − 0.000111 × B P D 2 × A C$
14	Campbell^[17]	$l n (W) = (− 4.564 + 0.282 × A C − 0.00331 × A C 2) + 6.9078$
15	Stirnemann^[18]	$l n (W) = 5.084820 − 54.06633 × (A C 100) 3 − 95.80076 × (A C 100 ) 3 × l n (A C 100 ) + 3.136370 × (H C 100 )$

序号	作者	计算公式
1	Hadlock I^[14]	$lg (W) = 1.3596 + 0.0064 × H C + 0.0424 × A C + 0.174 × F L + 0.00061 × B P D × A C − 0.00386 × A C × F L$
2	Hadlock Ⅳ^[14]	$lg (W) = 1.326 − 0.00326 × A C × F L + 0.0107 × H L + 0.0438 × A C + 0.158 × F L$
3	Hadlock Ⅱ^[14]	$lg (W) = 1.304 + 0.05281 × A C + 0.1938 × F L − 0.004 × A C × F L$
4	Hadlock Ⅲ^[14]	$lg (W) = 1.335 − 0.0034 × A C × F L + 0.0316 × B P D + 0.0457 × A C + 0.1623 × F L$
5	Ott^[14]	$lg (W) = 0.04355 × H C + 0.05394 × A C − 0.0008582 × H C × A C + 1.2504 × F L A C + 0.9339$
6	Combs^[14]	$lg (W) = 0.23718 × A C 2 × F L + 0.03312 × H C 2$
7	Bernstein^[15]	$lg (W) = 1.42482 + 0.01165 × H C + 0.03949 × A C − 0.00028 × A C 2 + 0.14147 × F L − 0.00662 × F L 2$
8	Hadlock V^[14]	$lg (W) = 1.5662 − 0.0108 H C + 0.0468 × A C + 0.171 × F L + 0.00034 × H C 2 − 0.003685 × A C × F L$
9	Jordaan^[16]	$lg (W) = 2.3231 + 0.02904 × A C + 0.0079 × H C − 0.0058 × B P D$
10	ShepardⅡ^[14]	$lg (W) = 1.2508 + 0.166 × B P D + 0.046 × A C − 0.002646 × A C × B P D$
11	Johnson^[15]	$lg (W) = 1.21633 + 0.06076 × H C − 0.00075 × A C 2 + 0.02107 × A C + 0.05261 × F L$
12	Merz^[14]	$W = − 3200.40479 + 57.07186 × A C + 15.90391 × B P D 2$
13	Warsof^[14]	$lg (W) = 1.467 + 0.144 × B P D + 0.032 × A C − 0.000111 × B P D 2 × A C$
14	Campbell^[17]	$l n (W) = (− 4.564 + 0.282 × A C − 0.00331 × A C 2) + 6.9078$
15	Stirnemann^[18]	$l n (W) = 5.084820 − 54.06633 × (A C 100) 3 − 95.80076 × (A C 100 ) 3 × l n (A C 100 ) + 3.136370 × (H C 100 )$

超参数	Logistic	CatBoostr	LightGBM	XGBoost
损失函数	-	-	gbdt	gbtree
提升器类型	reg:gamma	RMSE	regression	reg:gamma
树的最大深度	-	3	3	100
正则项系数	1	2	l2	2
训练样本采样率	-	0.7	0.9	0.7
特征采样率	-	-	0.8	0.7
随机种子	200	200	200	1000
分裂阈值	-	-	0.05	0.1

基本信息	均值	变异系数	数据范围
孕妇年龄（岁）	30.92±4.22	0.13	19~46
孕次	2（1，3）	0.60	1~10
产次	1（0，1）	1.23	0~5
孕前体重（kg）	55.00±8.75	0.15	35.0~96.5
孕期增重（kg)	12.75±4.12	0.32	-2.7~29.0
身高（cm）	157.78±5.59	0.03	137~176

参数	均数±标准差	变异系数	数据范围
GA（d）	88.66±4.51	0.05	69~100
CRL（cm）	6.34±0.70	0.11	4.5~8.4
NT（cm）	0.16±0.08	0.52	0.08~0.35
FHR	161.12±10.27	0.06	133~188
MVP（cm）	3.61±0.73	0.20	1.5~6.0

参数	均数±标准差	变异系数	数据范围
GA（d）	158.49±5.63	0.03	132~187
BPD（cm）	5.37±0.30	0.05	4.47~6.81
HC（cm）	19.99±0.94	0.04	17.36~27.90
AC（cm）	32.66±5.34	16.34	14.69~21.70
FL（cm）	3.83±0.94	0.24	3.09~4.70
HL（cm）	3.65±1.19	0.32	2.39~4.50
CER（cm）	2.33±0.13	0.05	1.89~2.95
MVP（cm）	4.21±1.26	0.29	2.23~8.20
UA Vmax（cm/s）	32.24±6.68	0.20	14.39~60.19
UA Vmin（cm/s）	10.04±2.43	0.24	3.80~32.72
UA S/D	3.26±0.62	0.19	0.80~13.11
UA RI	0.67±0.08	0.12	0.30~2.70
UA PI	1.14±0.18	0.16	0.46~1.88
FHR	149.28±7.72	0.05	102~178
MCA Vmax（cm/s）	27.80±5.63	0.20	12.06~46.16
MCA Vmin（cm/s）	5.85±1.58	0.27	2.49~15.60
MCA S/D	4.55±1.07	0.23	2.25~10.47
MCA RI	0.77±0.29	0.37	0.43~0.89
MCA PI	1.58±0.31	0.20	0.73~2.49

参数	均数±标准差	变异系数	数据范围
GA（d）	210.28±8.87	0.04	168~240
BPD（cm）	7.64±0.39	0.05	6.44~8.90
HC（cm）	27.72±1.41	0.05	24.10~31.48
AC（cm）	25.49±1.53	0.06	20.80~31.50
FL（cm）	5.67±0.62	0.11	4.69~6.59
MVP（cm）	4.29±0.88	0.20	2.19~9.00
UA Vmax(cm/s)	40.97±12.61	0.30	16.46~75.40
UA Vmin(cm/s)	15.40±4.38	0.28	6.43~31.81
UA S/D	2.72±0.45	0.16	1.40~4.55
UA RI	0.67±0.07	0.11	0.30~0.90
UA PI	0.96±0.26	0.27	0.50~1.53
FHR	144.13±7.92	0.05	113~174
MCA Vmax(cm/s)	37.78±8.32	0.22	14.00~63.21
MCA Vmin(cm/s)	7.14±2.56	0.36	2.10~20.00
MCA S/D	5.42±1.49	0.27	1.70~10.20
MCA RI	0.79±0.09	0.11	0.40~0.98
MCA PI	1.79±0.29	0.16	0.60~2.71

参数	均数±标准差	变异系数	数据范围
GA（d）	270.04±10.76	0.03	218~290
BPD（cm）	9.28±2.13	0.23	7.40~10.46
HC（cm）	32.95±8.23	0.25	26.97~36.60
AC（cm）	33.31±1.68	0.05	21.70~39.50
FL（cm）	7.27±2.01	0.27	5.79~8.00
MVP（cm）	4.17±1.77	0.42	0.90~9.60
UA Vmax（cm/s）	46.44±11.79	0.25	19.70~89.50
UA Vmin（cm/s）	20.72±5.89	0.28	7.36~44.60
UA S/D	2.27±0.39	0.17	1.40~4.190
UA RI	0.55±0.06	0.11	0.30~0.80
UA PI	0.83±0.14	0.17	0.40~1.22
FHR	141.70±9.71	0.06	111~189

资料	均数±标准差	变异系数	数据范围
分娩孕龄（d）	275.17±8.66	0.03	219~292
新生儿体重（g）	3167.90±402.60	0.12	1200~4500

References 28

1	Lubrano C, Taricco E, Coco C, et al. Perinatal and neonatal outcomes in fetal growth restriction and small for gestational age[J]. J Clin Med, 2022, 11(10): 2729.
2	Lu Y, Fu X, Chen F, et al. Prediction of fetal weight at varying gestational age in the absence of ultrasound examination using ensemble learning[J]. Artif Intell Med, 2020, 102: 101748.
3	Ma J, Cheng D, Zhang Z, et al. Evaluating the accuracy of sonographic fetal weight estimations using the Hadlock IV formula in a Chinese population[J]. Quant Imaging Med Surg, 2023, 13(6): 3726-3734.
4	Pu B, Li K, Li S, et al. Automatic fetal ultrasound standard plane recognition based on deep learning and ⅡoT[J]. IEEE Transactions on Industrial Informatics, 2021, 17(11): 7771-7780.
5	Pu B, Li K, Chen J, et al. HFSCCD: A hybrid neural network for fetal standard cardiac cycle detection in ultrasound videos[J]. IEEE J Biomed Health Inform, 2024, 28(5): 2943-2954.
6	He J, Yang L, Liang B, et al. Fetal cardiac ultrasound standard section detection model based on multitask learning and mixed attention mechanism[J]. Neurocomputing, 2024, 579: 127443.
7	石智红,李胜利. 人工智能在产前超声中的应用和研究进展[J/OL]. 中华医学超声杂志 (电子版), 2023, 20(1): 113-117.
8	谢子勋, 陈億龙, 宋维, 等. 融合科学智算的高危产妇妊娠结果评估模型及其应用研究[J]. 人工智能, 2024, (5): 109-125.
9	李昆, 柴玉梅, 赵红领, 等. 基于深度神经网络的胎儿体重预测[J]. 计算机科学, 2016, 43(z2): 73-76,82.
10	Caramês ETS, de Moraes-Neto VF, Bertozzi BG, et al. Identification of Fusarium sambucinum species complex by surface-enhanced Raman spectroscopy and XGBoost algorithm[J]. Food Chem, 2025, 480: 143848.
11	中华医学会超声医学分会妇产超声学组, 国家卫生健康委妇幼司全国产前诊断专家组医学影像组. 超声产前筛查指南 [J]. 中华超声影像学杂志, 2022, 31(1): 1-12.
12	任芸芸, 谢洪宁, 杨敏, 等. 中国产科超声检查指南[M].北京: 人民卫生出版社, 2019.
13	中华预防医学会出生缺陷预防与控制专业委员会产前超声诊断学组, 中华医学会超声医学分会妇产学组. 胎儿生物学参数超声参考值专家共识: 估测胎儿体重[J/OL]. 中华医学超声杂志(电子版), 2024, 21(8): 762-765.
14	Hammami A, Mazer Zumaeta A, Syngelaki A, et al. Ultrasonographic estimation of fetal weight: development of new model and assessment of performance of previous models[J]. Ultrasound Obstet Gynecol, 2018, 52(1): 35-43.
15	Lanowski JS, Lanowski G, Schippert C, et al. Ultrasound versus clinical examination to estimate fetal weight at term[J]. Geburtshilfe Frauenheilkd, 2017, 77(3): 276-283.
16	Eze CU, Egwuanumku KI, Agwuna KK, et al. Validity of common ultrasound methods of fetal weight estimation in late pregnancy among women in Kwale, Niger Delta region, Nigeria[J]. Afr Health Sci, 2015, 15(1): 206-216.
17	Campbell WA, Vintzileos AM, Neckles S, et al. Use of the femur length to estimate fetal weight in premature infants: preliminary results[J]. J Ultrasound Med, 1985, 4(11): 583-590.
18	Stirnemann J, Villar J, Salomon LJ, et al. International estimated fetal weight standards of the INTERGROWTH-21(st) Project[J]. Ultrasound Obstet Gynecol, 2017, 49(4): 478-486.
19	De Myttenaere A, Golden B, Le Grand B, et al. Using the mean absolute percentage error for regression models[DB/OL]. (2015-06-12)[2025-06-07].
20	林丽辉, 李国香, 刘良刚, 等. 应用超声胎儿测量系统预测胎儿体重的符合率研究[J]. 影像研究与医学应用, 2021, 5(15): 34-35.
21	王海红, 韩冰, 张慧. 影响超声估计胎儿体重准确性因素的研究进展[J]. 中华围产医学杂志, 2018, 21(10): 706-711.
22	Nash CM, Woolcott C, O'connell C, et al. Optimal timing of prenatal ultrasound in predicting birth weight in diabetic pregnancies[J]. J Obstet Gynaecol Can, 2020, 42(1): 48-53.
23	Dias T, Thilaganathan B. Is first-trimester crown-rump length associated with birthweight?[J]. BJOG, 2012, 119(3): 380.
24	Wang B, Zhang C. Risk evaluation of fetal growth restriction by combined screening in early and mid-pregnancy[J]. Pak J Med Sci, 2020, 36(7): 1708-1713.
25	蒋雯音.机器学习方法在早产和低出生体重儿预测中的应用[J]. 医学信息学杂志, 2019, 40(4): 59-63.
26	Plotka SS, Grzeszczyk MK, Szenejko PI, et al. Deep learning for estimation of fetal weight throughout the pregnancy from fetal abdominal ultrasound[J]. Am J Obstet Gynecol MFM, 2023, 5(12): 101182.
27	郭长胜. 胎儿宫内发育迟缓的危险因素及孕期运动对其干预作用的研究进展[J]. 儿童保健杂志, 2022, 30(10): 1091.
28	Ishrat S, Rahman MW, Rahman MR, et al. Leptin concentrations in maternal and umbilical cord blood in relation to maternal weight, birth weight and weight of the placenta[J]. Bangladesh Journal of Obstetrics & Gynaecology, 2008, 23(1): 3-7.

[1]	Yao Jiang, Cheng Jiang, Xiang Yu, Ying Tan, Xin Wen, Huiying Wen, Guiyan Peng, Shengli Li. Performance of an attention-enhanced multi-task model for uterine anatomical structure detection and segmentation [J]. Chinese Journal of Medical Ultrasound (Electronic Edition), 2025, 22(08): 703-710.
[2]	Minglang Chen, Kai Xu, Zhixi Huang, Bocheng Liang, Jie He, Haishan Huang, Weibo Ma, Ying Tan, Zhiying Zou, Xiaotang Liu, Guiyan Peng, Jiaxi Chen, Xiaohong Zhong. MobileNetV4: an intelligent tool for diagnosis of aortic arch branching anomalies based on prenatal ultrasound images [J]. Chinese Journal of Medical Ultrasound (Electronic Edition), 2025, 22(08): 711-720.
[3]	Xiaofang Fu, Qinghan Yang, Changqin Sun, Mengjie Dou, Junpu Hu, Hao Sun, Faqin Lyu. Clinical value of a YOLO 11-based ultrasound detection model for limb long bone fracture ends [J]. Chinese Journal of Medical Ultrasound (Electronic Edition), 2025, 22(06): 541-546.
[4]	Guannan He, Ying Tan, Yuhuan Lu, Bin Pu, Shuihua Yang, Rentie Zhang, Ming Chen, Zhihong Shi, Xiaohong Zhong, Xi Chen, Liuyi Yan, Shengli Li. Application of artificial intelligence in quality control of standard views for fetal echocardiography: a multi-center study [J]. Chinese Journal of Medical Ultrasound (Electronic Edition), 2025, 22(05): 388-396.
[5]	Yan Wang, Xiaohang Zhang, Suzhen Ran, Chunyan Zhong, Jinwei Zhang, Xi Wang. Prenatal ultrasonographic features of twin anemia-polycythemia sequence: analysis of 26 cases [J]. Chinese Journal of Medical Ultrasound (Electronic Edition), 2025, 22(05): 462-469.
[6]	Yin Chen, Ying Tan, Bohan Tan, Guannan He, Lei Wang, Xin Wen, Qiaozhen Zhu, Bocheng Liang, Shengli1 Li. Intelligent quality assessment and diagnosis of fetal omphalocele using YOLO V8 in ultrasound imaging [J]. Chinese Journal of Medical Ultrasound (Electronic Edition), 2025, 22(04): 305-310.
[7]	Meng Dong, Xinlu Wang, Guangyu Zhu, Xin Geng, Hua Yang. Efficacy of artificial intelligence-assisted pelvic floor ultrasound in diagnosing cystocele [J]. Chinese Journal of Obstetrics & Gynecology and Pediatrics(Electronic Edition), 2025, 21(03): 296-303.
[8]	Renjie Zheng, Zunshu Zhang, Chen Huang. Application and challenges of artificial intelligence-based pathomics in the diagnosis and management of gastric cancer [J]. Chinese Archives of General Surgery(Electronic Edition), 2025, 19(04): 274-280.
[9]	Shiwang Huang, Yunkai Qie, Hailong Hu. Application of artificial intelligence in non-invasive urine-based diagnostic methods for bladder cancer [J]. Chinese Journal of Endourology(Electronic Edition), 2025, 19(03): 283-287.
[10]	Yuming Zhan, Xiang Zhang, Shangeng Weng. Research progress of artificial intelligence in the precise diagnosis and treatment of retroperitoneal tumors [J]. Chinese Journal of Hernia and Abdominal Wall Surgery(Electronic Edition), 2025, 19(04): 371-376.
[11]	Yanyun Zhang, Qizhi Bai, Jianwei Zhang, Dazhi Guo, Dan Ma, Mingfu Gong. AI-based CT quantitative analysis was used for efficacy analysis in the evaluation of pulmonary nodules and degree of infiltration [J]. Chinese Journal of Lung Diseases(Electronic Edition), 2025, 18(03): 411-415.
[12]	Yuanyuan Li, Rongshan Li. Machine learning: a new frontier in research and therapy of kidney diseases [J]. Chinese Journal of Kidney Disease Investigation(Electronic Edition), 2025, 14(04): 181-187.
[13]	Huan Chen, Menghan Li, Weihong Yu. Pay attention to the application of artificial intelligence in constructing a three-tier screening and diagnosis-treatment system for ocular fundus diseases [J]. Chinese Journal of Ophthalmologic Medicine(Electronic Edition), 2025, 15(03): 129-134.
[14]	Lingjie Wang, Aiping Wang, Chaojun Li, Yueyou Ding, Deye Yang, Qing Zhao, Zhaoqiang Cui, Jingkun Wang, Hongyu Wang. Expert Consensus on Research and Development Strategies for Innovative Heart and Vascular Health Technologies (2024, Shanghai) [J]. Chinese Journal of Clinicians(Electronic Edition), 2025, 19(05): 323-336.
[15]	Ye Ma, Hang Qu, Wei Wang. Role of artificial intelligence in evaluation of cerebrovascular disease imaging [J]. Chinese Journal of Clinicians(Electronic Edition), 2025, 19(03): 211-215.

Please choose a citation manager

Content to export