The needs of dentists for artificial intelligence models to read pediatric panoramic radiographs
AIM:In recent years,the research on artificial intelligence(AI)in the field of pediatric dentistry has progressively advanced.Most of the research materials are imaging data,among which panoramic radiographs are prevalent.Thongh functions such as tooth detection and physiological age inference could be achieved,with the focus on constructing the corresponding models,many difficulties still remain to be addressed before clinical application.The aim of this study is to understand the attitude and requirements of dentists towards AI models to read pediatric panoramic radiographs in practical operations,aiding for the design and transformation of related research.METHODS:A questionnaire survey was conducted among 600 dentists from 30 provinces,autonomous regions and municipalities in China.In order to verify the consistency of the investigation results and the actual clinical work,the description contents from single-center pediatric panoramic tomography reports were collected and analyzed.RESULTS:The majority of dentists surveyed were interested in and confident about the future of AI,although the minority worried about to be replaced.The most required function was to assess the overall development of tooth development,followed by the generation of reviewable text reports.Focus on reading pediatric panoramic radiographs of dentists was the abnormality of the number of teeth,and then the abnormality of tooth eruption and loss.The most common difficulty was the identification of permanent tooth germs affected by structural overlap,and the second was missing other problems beyond chief complaint.CONCLUSION:AI was welcome among Chinese dentists,and the development of AI-aided software to read pediatric panoramic radiographs should take the actual needs of dentists into account.
artificial intelligencechildrenpanoramic radiographysurveys and questionnaires
NETL
NSTL
万方数据
