{"title":"测序验证了基于ehr的儿科患者努南综合征检测的深度学习模型。","authors":"Zeyu Yang, Amy Shikany, Ammar Husami, Xinjian Wang, Eneida Mendonca, K Nicole Weaver, Jing Chen","doi":"10.1038/s41525-025-00512-5","DOIUrl":null,"url":null,"abstract":"<p><p>Despite advanced diagnostic tools, early detection of rare genetic conditions like Noonan syndrome (NS) remains challenging. We evaluated a deep learning model's real-world performance in identifying potential NS cases using electronic health record (EHR) data, validated through genetic sequencing and clinical assessment. The model analyzed 92,428 patients, identifying 171 high-risk individuals (score > 0.8) who underwent comprehensive review. Among these, 86 had prior genetic diagnoses, including three NS cases diagnosed during the study period. Genetic sequencing of remaining patients identified two additional NS cases with pathogenic variants. The model achieved 2.92% precision and 99.82% specificity. While precision was lower than prior validation (33.3%), this reflected expected differences in disease prevalence rather than model degradation. NS-associated phenotypes were enriched among high-risk patients, and trajectory analysis showed potential for earlier identification, highlighting both promise and limitations of EHR-based computational screening tools.</p>","PeriodicalId":19273,"journal":{"name":"NPJ Genomic Medicine","volume":"10 1","pages":"56"},"PeriodicalIF":4.7000,"publicationDate":"2025-07-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12280026/pdf/","citationCount":"0","resultStr":"{\"title\":\"Sequencing validates deep learning models for EHR-based detection of Noonan syndrome in pediatric patients.\",\"authors\":\"Zeyu Yang, Amy Shikany, Ammar Husami, Xinjian Wang, Eneida Mendonca, K Nicole Weaver, Jing Chen\",\"doi\":\"10.1038/s41525-025-00512-5\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>Despite advanced diagnostic tools, early detection of rare genetic conditions like Noonan syndrome (NS) remains challenging. We evaluated a deep learning model's real-world performance in identifying potential NS cases using electronic health record (EHR) data, validated through genetic sequencing and clinical assessment. The model analyzed 92,428 patients, identifying 171 high-risk individuals (score > 0.8) who underwent comprehensive review. Among these, 86 had prior genetic diagnoses, including three NS cases diagnosed during the study period. Genetic sequencing of remaining patients identified two additional NS cases with pathogenic variants. The model achieved 2.92% precision and 99.82% specificity. While precision was lower than prior validation (33.3%), this reflected expected differences in disease prevalence rather than model degradation. NS-associated phenotypes were enriched among high-risk patients, and trajectory analysis showed potential for earlier identification, highlighting both promise and limitations of EHR-based computational screening tools.</p>\",\"PeriodicalId\":19273,\"journal\":{\"name\":\"NPJ Genomic Medicine\",\"volume\":\"10 1\",\"pages\":\"56\"},\"PeriodicalIF\":4.7000,\"publicationDate\":\"2025-07-21\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12280026/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"NPJ Genomic Medicine\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1038/s41525-025-00512-5\",\"RegionNum\":2,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"GENETICS & HEREDITY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"NPJ Genomic Medicine","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1038/s41525-025-00512-5","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"GENETICS & HEREDITY","Score":null,"Total":0}
Sequencing validates deep learning models for EHR-based detection of Noonan syndrome in pediatric patients.
Despite advanced diagnostic tools, early detection of rare genetic conditions like Noonan syndrome (NS) remains challenging. We evaluated a deep learning model's real-world performance in identifying potential NS cases using electronic health record (EHR) data, validated through genetic sequencing and clinical assessment. The model analyzed 92,428 patients, identifying 171 high-risk individuals (score > 0.8) who underwent comprehensive review. Among these, 86 had prior genetic diagnoses, including three NS cases diagnosed during the study period. Genetic sequencing of remaining patients identified two additional NS cases with pathogenic variants. The model achieved 2.92% precision and 99.82% specificity. While precision was lower than prior validation (33.3%), this reflected expected differences in disease prevalence rather than model degradation. NS-associated phenotypes were enriched among high-risk patients, and trajectory analysis showed potential for earlier identification, highlighting both promise and limitations of EHR-based computational screening tools.
NPJ Genomic MedicineBiochemistry, Genetics and Molecular Biology-Molecular Biology
CiteScore
9.40
自引率
1.90%
发文量
67
审稿时长
17 weeks
期刊介绍:
npj Genomic Medicine is an international, peer-reviewed journal dedicated to publishing the most important scientific advances in all aspects of genomics and its application in the practice of medicine.
The journal defines genomic medicine as "diagnosis, prognosis, prevention and/or treatment of disease and disorders of the mind and body, using approaches informed or enabled by knowledge of the genome and the molecules it encodes." Relevant and high-impact papers that encompass studies of individuals, families, or populations are considered for publication. An emphasis will include coupling detailed phenotype and genome sequencing information, both enabled by new technologies and informatics, to delineate the underlying aetiology of disease. Clinical recommendations and/or guidelines of how that data should be used in the clinical management of those patients in the study, and others, are also encouraged.