{"title":"Clinical validation of a deep-learning-based bone age software in healthy Korean children.","authors":"Hyo-Kyoung Nam, Winnah Wu-In Lea, Zepa Yang, Eunjin Noh, Young-Jun Rhie, Kee-Hyoung Lee, Suk-Joo Hong","doi":"10.6065/apem.2346050.025","DOIUrl":null,"url":null,"abstract":"<p><strong>Purpose: </strong>Bone age (BA) is needed to assess developmental status and growth disorders. We evaluated the clinical performance of a deep-learning-based BA software to estimate the chronological age (CA) of healthy Korean children.</p><p><strong>Methods: </strong>This retrospective study included 371 healthy children (217 boys, 154 girls), aged between 4 and 17 years, who visited the Department of Pediatrics for health check-ups between January 2017 and December 2018. A total of 553 left-hand radiographs from 371 healthy Korean children were evaluated using a commercial deep-learning-based BA software (BoneAge, Vuno, Seoul, Korea). The clinical performance of the deep learning (DL) software was determined using the concordance rate and Bland-Altman analysis via comparison with the CA.</p><p><strong>Results: </strong>A 2-sample t-test (P<0.001) and Fisher exact test (P=0.011) showed a significant difference between the normal CA and the BA estimated by the DL software. There was good correlation between the 2 variables (r=0.96, P<0.001); however, the root mean square error was 15.4 months. With a 12-month cutoff, the concordance rate was 58.8%. The Bland-Altman plot showed that the DL software tended to underestimate the BA compared with the CA, especially in children under the age of 8.3 years.</p><p><strong>Conclusion: </strong>The DL-based BA software showed a low concordance rate and a tendency to underestimate the BA in healthy Korean children.</p>","PeriodicalId":44915,"journal":{"name":"Annals of Pediatric Endocrinology & Metabolism","volume":null,"pages":null},"PeriodicalIF":2.8000,"publicationDate":"2024-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11076234/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Annals of Pediatric Endocrinology & Metabolism","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.6065/apem.2346050.025","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/1/24 0:00:00","PubModel":"Epub","JCR":"Q3","JCRName":"ENDOCRINOLOGY & METABOLISM","Score":null,"Total":0}
引用次数: 0
Abstract
Purpose: Bone age (BA) is needed to assess developmental status and growth disorders. We evaluated the clinical performance of a deep-learning-based BA software to estimate the chronological age (CA) of healthy Korean children.
Methods: This retrospective study included 371 healthy children (217 boys, 154 girls), aged between 4 and 17 years, who visited the Department of Pediatrics for health check-ups between January 2017 and December 2018. A total of 553 left-hand radiographs from 371 healthy Korean children were evaluated using a commercial deep-learning-based BA software (BoneAge, Vuno, Seoul, Korea). The clinical performance of the deep learning (DL) software was determined using the concordance rate and Bland-Altman analysis via comparison with the CA.
Results: A 2-sample t-test (P<0.001) and Fisher exact test (P=0.011) showed a significant difference between the normal CA and the BA estimated by the DL software. There was good correlation between the 2 variables (r=0.96, P<0.001); however, the root mean square error was 15.4 months. With a 12-month cutoff, the concordance rate was 58.8%. The Bland-Altman plot showed that the DL software tended to underestimate the BA compared with the CA, especially in children under the age of 8.3 years.
Conclusion: The DL-based BA software showed a low concordance rate and a tendency to underestimate the BA in healthy Korean children.
期刊介绍：
The Annals of Pediatric Endocrinology & Metabolism Journal is the official publication of the Korean Society of Pediatric Endocrinology. Its formal abbreviated title is “Ann Pediatr Endocrinol Metab”. It is a peer-reviewed open access journal of medicine published in English. The journal was launched in 1996 under the title of ‘Journal of Korean Society of Pediatric Endocrinology’ until 2011 (pISSN 1226-2242). Since 2012, the title is now changed to ‘Annals of Pediatric Endocrinology & Metabolism’. The Journal is published four times per year on the last day of March, June, September, and December. It is widely distributed for free to members of the Korean Society of Pediatric Endocrinology, medical schools, libraries, and academic institutions. The journal is indexed/tracked/covered by web sites of PubMed Central, PubMed, Emerging Sources Citation Index (ESCI), Scopus, EBSCO, EMBASE, KoreaMed, KoMCI, KCI, Science Central, DOI/CrossRef, Directory of Open Access Journals(DOAJ), and Google Scholar. The aims of Annals of Pediatric Endocrinology & Metabolism are to contribute to the advancements in the fields of pediatric endocrinology & metabolism through the scientific reviews and interchange of all of pediatric endocrinology and metabolism. It aims to reflect the latest clinical, translational, and basic research trends from worldwide valuable achievements. In addition, genome research, epidemiology, public education and clinical practice guidelines in each country are welcomed for publication. The Journal particularly focuses on research conducted with Asian-Pacific children whose genetic and environmental backgrounds are different from those of the Western. Area of specific interest include the following : Growth, puberty, glucose metabolism including diabetes mellitus, obesity, nutrition, disorders of sexual development, pituitary, thyroid, parathyroid, adrenal cortex, bone or other endocrine and metabolic disorders from infancy through adolescence.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
