A comparison of two artificial intelligence-based methods for assessing bone age in Turkish children: BoneXpert and VUNO Med-Bone Age

• DOI: 10.4274/dir.2024.242790
• Publication Date: 2024-09-02T12:18:15Z
• Authors: Evrim Özmen; Hande Özen Atalay; Evren Uzer; Mert Veznikli

This study aimed to evaluate the validity of two artificial intelligence (AI)-based bone age assessment programs, BoneXpert and VUNO Med-Bone Age (VUNO), compared with manual assessments using Greulich-Pyle method in Turkish children. METHODSThis included a cohort 292 pediatric cases, ranging from 1 15 years an equal gender number distribution each group.Two radiologists, who were unaware determined by AI, independently evaluated age.The statistical involved intraclass correlation coefficient (ICC) measure level agreement between AI-based assessments. RESULTSThe ICC coefficients for measurements radiologists indicate almost perfect agreement.When all regardless group, analyzed, nearly positive was observed software measurements.When calculations separated analyzed separately girls boys, there no statistically significant difference methods boys; however, 0.990 0.982 calculated BoneXpert, respectively, this 0.008 (z = 2.528, P 0.012) girls.Accordingly, showed higher BoneXpert.The agreements demonstrated packages prepubescent group much more pronounced boys.After 8 9 both AI equal. CONCLUSIONBoth high assessing age.Furthermore, has prepubertal girls.These results suggest that may be slightly effective determining age, indicating its potential as highly reliable tool CLINICAL SIGNIFICANCEInvestigating most suitable program population could clinically significant.