Noninvasive early prediction of preeclampsia in pregnancy using retinal vascular features

Yuxuan Wu; Lixia  Shen; Lanqin  Zhao; Xiaohong  Lin; Miaohong  Xu; Zhenjun  Tu; Yihong  Huang; Lingyi  Kong; Zhenzhe  Lin; Duoru  Lin; Lixue  Liu; Xun  Wang; Zizheng  Cao; Xi  Chen; Shengmei  Zhou; Weiling  Hu; Yunjian  Huang; Shiyuan  Chen; Meimei  Dongye; Xulin  Zhang; Dongni  Wang; Danli Shi; Zilian  Wang; Xiaohang  Wu; Dongyu  Wang; Haotian  Lin

doi:10.1038/s41746-025-01582-6

Noninvasive early prediction of preeclampsia in pregnancy using retinal vascular features

Yuxuan Wu
, Lixia Shen
, Lanqin Zhao
, Xiaohong Lin
, Miaohong Xu
, Zhenjun Tu
, Yihong Huang
, Lingyi Kong
, Zhenzhe Lin
, Duoru Lin
, Lixue Liu
, Xun Wang
, Zizheng Cao
, Xi Chen
, Shengmei Zhou
, Weiling Hu
, Yunjian Huang
, Shiyuan Chen
, Meimei Dongye
, Xulin Zhang

Dongni Wang, Danli Shi (Corresponding Author), Zilian Wang (Corresponding Author), Xiaohang Wu (Corresponding Author), Dongyu Wang (Corresponding Author), Haotian Lin (Corresponding Author)

Research output: Journal article publication › Journal article › Academic research › peer-review

7 Citations (Scopus)

Abstract

Preeclampsia (PE), a severe hypertensive disorder during pregnancy, significantly contributes to maternal and neonatal mortality. Existing prediction biomarkers are often invasive and expensive, hindering their widespread application. This study introduces PROMPT (Preeclampsia Risk factor + Ophthalmic data + Mean arterial pressure Prediction Test), an AI-driven model leveraging retinal photography for PE prediction, registered at ChiCTR (ChiCTR2100049850) in August 2021. Analyzing 1812 pregnancies before 14 gestational weeks, we extracted retinal parameters using a deep learning system. The PROMPT achieved an AUC of 0.87 (0.83–0.90) for PE prediction and 0.91 (0.85–0.97) for preterm PE prediction using machine learning, significantly outperforming the baseline model (p < 0.001). It also improved detection of severe adverse pregnancy outcomes from 35% to 41%. Economically, PROMPT was estimated to avert 1809 PE cases and saved over $50 million per 100,000 screenings. These results position PROMPT as a non-invasive and cost-effective tool for prenatal care, especially valuable in low- and middle-income countries.

Original language	English
Article number	188
Pages (from-to)	1-9
Number of pages	9
Journal	npj Digital Medicine
Volume	8
Issue number	1
DOIs	https://doi.org/10.1038/s41746-025-01582-6
Publication status	Published - 5 Apr 2025

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

More information

10.1038/s41746-025-01582-6

http://hdl.handle.net/10397/112833

Cite this

Wu, Y., Shen, L., Zhao, L., Lin, X., Xu, M., Tu, Z., Huang, Y., Kong, L., Lin, Z., Lin, D., Liu, L., Wang, X., Cao, Z., Chen, X., Zhou, S., Hu, W., Huang, Y., Chen, S., Dongye, M., ... Lin, H. (2025). Noninvasive early prediction of preeclampsia in pregnancy using retinal vascular features. npj Digital Medicine, 8(1), 1-9. Article 188. https://doi.org/10.1038/s41746-025-01582-6

@article{f6b34b1c095c49fca738e4c17ce5c6c4,

title = "Noninvasive early prediction of preeclampsia in pregnancy using retinal vascular features",

abstract = "Preeclampsia (PE), a severe hypertensive disorder during pregnancy, significantly contributes to maternal and neonatal mortality. Existing prediction biomarkers are often invasive and expensive, hindering their widespread application. This study introduces PROMPT (Preeclampsia Risk factor + Ophthalmic data + Mean arterial pressure Prediction Test), an AI-driven model leveraging retinal photography for PE prediction, registered at ChiCTR (ChiCTR2100049850) in August 2021. Analyzing 1812 pregnancies before 14 gestational weeks, we extracted retinal parameters using a deep learning system. The PROMPT achieved an AUC of 0.87 (0.83–0.90) for PE prediction and 0.91 (0.85–0.97) for preterm PE prediction using machine learning, significantly outperforming the baseline model (p < 0.001). It also improved detection of severe adverse pregnancy outcomes from 35\% to 41\%. Economically, PROMPT was estimated to avert 1809 PE cases and saved over \$50 million per 100,000 screenings. These results position PROMPT as a non-invasive and cost-effective tool for prenatal care, especially valuable in low- and middle-income countries.",

author = "Yuxuan Wu and Lixia Shen and Lanqin Zhao and Xiaohong Lin and Miaohong Xu and Zhenjun Tu and Yihong Huang and Lingyi Kong and Zhenzhe Lin and Duoru Lin and Lixue Liu and Xun Wang and Zizheng Cao and Xi Chen and Shengmei Zhou and Weiling Hu and Yunjian Huang and Shiyuan Chen and Meimei Dongye and Xulin Zhang and Dongni Wang and Danli Shi and Zilian Wang and Xiaohang Wu and Dongyu Wang and Haotian Lin",

note = "Publisher Copyright: {\textcopyright} The Author(s) 2025.",

year = "2025",

month = apr,

day = "5",

doi = "10.1038/s41746-025-01582-6",

language = "English",

volume = "8",

pages = "1--9",

journal = "npj Digital Medicine",

issn = "2398-6352",

publisher = "Nature Publishing Group",

number = "1",

}

Wu, Y, Shen, L, Zhao, L, Lin, X, Xu, M, Tu, Z, Huang, Y, Kong, L, Lin, Z, Lin, D, Liu, L, Wang, X, Cao, Z, Chen, X, Zhou, S, Hu, W, Huang, Y, Chen, S, Dongye, M, Zhang, X, Wang, D, Shi, D, Wang, Z, Wu, X, Wang, D & Lin, H 2025, 'Noninvasive early prediction of preeclampsia in pregnancy using retinal vascular features', npj Digital Medicine, vol. 8, no. 1, 188, pp. 1-9. https://doi.org/10.1038/s41746-025-01582-6

TY - JOUR

T1 - Noninvasive early prediction of preeclampsia in pregnancy using retinal vascular features

AU - Wu, Yuxuan

AU - Shen, Lixia

AU - Zhao, Lanqin

AU - Lin, Xiaohong

AU - Xu, Miaohong

AU - Tu, Zhenjun

AU - Huang, Yihong

AU - Kong, Lingyi

AU - Lin, Zhenzhe

AU - Lin, Duoru

AU - Liu, Lixue

AU - Wang, Xun

AU - Cao, Zizheng

AU - Chen, Xi

AU - Zhou, Shengmei

AU - Hu, Weiling

AU - Huang, Yunjian

AU - Chen, Shiyuan

AU - Dongye, Meimei

AU - Zhang, Xulin

AU - Wang, Dongni

AU - Shi, Danli

AU - Wang, Zilian

AU - Wu, Xiaohang

AU - Wang, Dongyu

AU - Lin, Haotian

PY - 2025/4/5

Y1 - 2025/4/5

N2 - Preeclampsia (PE), a severe hypertensive disorder during pregnancy, significantly contributes to maternal and neonatal mortality. Existing prediction biomarkers are often invasive and expensive, hindering their widespread application. This study introduces PROMPT (Preeclampsia Risk factor + Ophthalmic data + Mean arterial pressure Prediction Test), an AI-driven model leveraging retinal photography for PE prediction, registered at ChiCTR (ChiCTR2100049850) in August 2021. Analyzing 1812 pregnancies before 14 gestational weeks, we extracted retinal parameters using a deep learning system. The PROMPT achieved an AUC of 0.87 (0.83–0.90) for PE prediction and 0.91 (0.85–0.97) for preterm PE prediction using machine learning, significantly outperforming the baseline model (p < 0.001). It also improved detection of severe adverse pregnancy outcomes from 35% to 41%. Economically, PROMPT was estimated to avert 1809 PE cases and saved over $50 million per 100,000 screenings. These results position PROMPT as a non-invasive and cost-effective tool for prenatal care, especially valuable in low- and middle-income countries.

AB - Preeclampsia (PE), a severe hypertensive disorder during pregnancy, significantly contributes to maternal and neonatal mortality. Existing prediction biomarkers are often invasive and expensive, hindering their widespread application. This study introduces PROMPT (Preeclampsia Risk factor + Ophthalmic data + Mean arterial pressure Prediction Test), an AI-driven model leveraging retinal photography for PE prediction, registered at ChiCTR (ChiCTR2100049850) in August 2021. Analyzing 1812 pregnancies before 14 gestational weeks, we extracted retinal parameters using a deep learning system. The PROMPT achieved an AUC of 0.87 (0.83–0.90) for PE prediction and 0.91 (0.85–0.97) for preterm PE prediction using machine learning, significantly outperforming the baseline model (p < 0.001). It also improved detection of severe adverse pregnancy outcomes from 35% to 41%. Economically, PROMPT was estimated to avert 1809 PE cases and saved over $50 million per 100,000 screenings. These results position PROMPT as a non-invasive and cost-effective tool for prenatal care, especially valuable in low- and middle-income countries.

UR - https://www.scopus.com/pages/publications/105002896663

U2 - 10.1038/s41746-025-01582-6

DO - 10.1038/s41746-025-01582-6

M3 - Journal article

SN - 2398-6352

VL - 8

SP - 1

EP - 9

JO - npj Digital Medicine

JF - npj Digital Medicine

IS - 1

M1 - 188

ER -

Noninvasive early prediction of preeclampsia in pregnancy using retinal vascular features

Abstract

UN SDGs

More information

Other files and links

Fingerprint

Cite this