Artificial intelligence for detecting fetal orofacial clefts and advancing medical education
Orofacial clefts are among the most common congenital craniofacial abnormalities, yet accurate prenatal detection remains challenging due to the scarcity of experienced specialists and the relative rarity of the condition. Early and reliable diagnosis is essential to enable timely clinical intervention and reduce associated morbidity. Here we show that an artificial intelligence system, trained on over 45,139 ultrasound images from 9,215 fetuses across 22 hospitals, can diagnose fetal orofacial clefts with sensitivity and specificity exceeding 93% and 95% respectively, matching the performance of senior radiologists and substantially outperforming junior radiologists. When used as a medical copilot, the system raises junior radiologists’ sensitivity by more than 6 %. Beyond direct diagnostic assistance, the system also accelerates the development of clinical expertise. A pilot study involving 24 radiologists and trainees demonstrated that the model can improve the expertise development for rare conditions. This dual-purpose approach offers a scalable solution for improving both diagnostic accuracy and specialist training in settings where experienced radiologists are scarce. Researchers developed an AI system matching senior radiologist accuracy in prenatal orofacial cleft diagnosis, while also serving as an educational tool. In a pilot study, AI-trained groups improved diagnostic performance over traditionally trained peers.
Authors
- Xin Yang (ORCID: https://orcid.org/0000-0003-4653-6524)
- Xiliang Zhu (ORCID: https://orcid.org/0000-0001-9351-3005)
- Guangzhi He (ORCID: https://orcid.org/0000-0003-1770-3522)
- Alejandro Federico Frangi (ORCID: https://orcid.org/0000-0002-2675-528X)
- Libing Liang (ORCID: https://orcid.org/0009-0004-3724-0435)
- Yuanji Zhang (ORCID: https://orcid.org/0000-0002-5398-7588)
- Dong Ni (ORCID: https://orcid.org/0000-0002-9146-6003)
- Xiaoqiong Huang (ORCID: https://orcid.org/0000-0003-2354-0758)
- Haoran Dou (ORCID: https://orcid.org/0000-0001-8628-5489)
- Linliang Yin (ORCID: https://orcid.org/0000-0002-6486-3564)
- LI Hong-long
- Jiewei Lai (ORCID: https://orcid.org/0000-0001-9872-9956)
- Zha Yang (ORCID: https://orcid.org/0009-0001-9265-5148)
- Xia Gao (ORCID: https://orcid.org/0009-0001-0480-1468)
- Liling Shi
- Yongsong Zhou (ORCID: https://orcid.org/0009-0000-2259-0070)
- Jialin Zhang
- Y Huang (ORCID: https://orcid.org/0009-0005-3406-4875)
- Siying Liang
- Mengqin Yuan
- Jiuping Li
- Yuhan Zhang
- Yan Cao
- Rui Li
- Cheng Yu
- Xuedong Deng (ORCID: https://orcid.org/0009-0007-9127-9711)
- Yi Xiong (ORCID: https://orcid.org/0009-0009-2535-8571)
- Chen Ling
- Wenxiong Pan (ORCID: https://orcid.org/0000-0001-7038-3455)
- Hongyu Zheng
- Chaoyu Chen
Institutions
- Shenzhen University (CN)
- Nanjing University of Information Science and Technology (CN)
- Manchester Academic Health Science Centre (GB)
- University of Manchester (GB)
- Shenzhen Metro (China) (CN)
- Northwest Women's and Children's Hospital (CN)
- Hangzhou Women’s Hospital (CN)
- The People's Hospital of Guangxi Zhuang Autonomous Region (CN)
- Suzhou Municipal Hospital (CN)
- Suzhou Traditional Chinese Medicine Hospital (CN)
- Shenzhen University Health Science Center (CN)
- Shenzhen Luohu People's Hospital (CN)
- Shanxi Provincial Children's Hospital (CN)
- Shenzhen Technology University (CN)
- NIHR Manchester Biomedical Research Centre (GB)
- Nanjing Medical University (CN)
- Manchester University (US)
- KU Leuven (BE)
Publication Details
- Journal
- Nature Communications
- Published
- 2026-06-16
- DOI
- https://doi.org/10.1038/s41467-026-74119-4
- Primary Topic
- Cleft Lip and Palate Research
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Institute for Health and Care Research
- National Natural Science Foundation of China