Factors influencing the development of artificial intelligence in orthodontics

IF 2.4 3区医学 Q2 DENTISTRY, ORAL SURGERY & MEDICINE

Orthodontics & Craniofacial Research Pub Date : 2024-05-07 DOI:10.1111/ocr.12806

Ju-Myung Lee, Jun-Ho Moon, Ji-Ae Park, Jong-Hak Kim, Shin-Jae Lee

{"title":"Factors influencing the development of artificial intelligence in orthodontics","authors":"Ju-Myung Lee, Jun-Ho Moon, Ji-Ae Park, Jong-Hak Kim, Shin-Jae Lee","doi":"10.1111/ocr.12806","DOIUrl":null,"url":null,"abstract":"<div>\n \n \n <section>\n \n <h3> Objectives</h3>\n \n <p>Since developing AI procedures demands significant computing resources and time, the implementation of a careful experimental design is essential. The purpose of this study was to investigate factors influencing the development of AI in orthodontics.</p>\n </section>\n \n <section>\n \n <h3> Materials and Methods</h3>\n \n <p>A total of 162 AI models were developed, with various combinations of sample sizes (170, 340, 679), input variables (40, 80, 160), output variables (38, 76, 154), training sessions (100, 500, 1000), and computer specifications (new vs. old). The TabNet deep-learning algorithm was used to develop these AI models, and leave-one-out cross-validation was applied in training. The goodness-of-fit of the regression models was compared using the adjusted coefficient of determination values, and the best-fit model was selected accordingly. Multiple linear regression analyses were employed to investigate the relationship between the influencing factors.</p>\n </section>\n \n <section>\n \n <h3> Results</h3>\n \n <p>Increasing the number of training sessions enhanced the effectiveness of the AI models. The best-fit regression model for predicting the computational time of AI, which included logarithmic transformation of time, sample size, and training session variables, demonstrated an adjusted coefficient of determination of 0.99.</p>\n </section>\n \n <section>\n \n <h3> Conclusion</h3>\n \n <p>The study results show that estimating the time required for AI development may be possible using logarithmic transformations of time, sample size, and training session variables, followed by applying coefficients estimated through several pilot studies with reduced sample sizes and reduced training sessions.</p>\n </section>\n </div>","PeriodicalId":19652,"journal":{"name":"Orthodontics & Craniofacial Research","volume":"27 S2","pages":"6-12"},"PeriodicalIF":2.4000,"publicationDate":"2024-05-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/epdf/10.1111/ocr.12806","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Orthodontics & Craniofacial Research","FirstCategoryId":"3","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1111/ocr.12806","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"DENTISTRY, ORAL SURGERY & MEDICINE","Score":null,"Total":0}

引用次数: 0

Abstract

Objectives

Since developing AI procedures demands significant computing resources and time, the implementation of a careful experimental design is essential. The purpose of this study was to investigate factors influencing the development of AI in orthodontics.

Materials and Methods

A total of 162 AI models were developed, with various combinations of sample sizes (170, 340, 679), input variables (40, 80, 160), output variables (38, 76, 154), training sessions (100, 500, 1000), and computer specifications (new vs. old). The TabNet deep-learning algorithm was used to develop these AI models, and leave-one-out cross-validation was applied in training. The goodness-of-fit of the regression models was compared using the adjusted coefficient of determination values, and the best-fit model was selected accordingly. Multiple linear regression analyses were employed to investigate the relationship between the influencing factors.

Results

Increasing the number of training sessions enhanced the effectiveness of the AI models. The best-fit regression model for predicting the computational time of AI, which included logarithmic transformation of time, sample size, and training session variables, demonstrated an adjusted coefficient of determination of 0.99.

Conclusion

The study results show that estimating the time required for AI development may be possible using logarithmic transformations of time, sample size, and training session variables, followed by applying coefficients estimated through several pilot studies with reduced sample sizes and reduced training sessions.

Abstract Image

查看原文本刊更多论文

影响正畸领域人工智能发展的因素。

目的：由于开发人工智能程序需要大量的计算资源和时间，因此实施精心的实验设计至关重要。本研究旨在调查影响正畸学人工智能发展的因素：共开发了 162 个人工智能模型，样本量（170、340、679）、输入变量（40、80、160）、输出变量（38、76、154）、训练次数（100、500、1000）和计算机规格（新与旧）的组合各不相同。这些人工智能模型的开发采用了 TabNet 深度学习算法，并在训练中进行了留一交叉验证。使用调整后的决定系数值对回归模型的拟合度进行比较，并据此选择最佳拟合模型。采用多元线性回归分析研究影响因素之间的关系：结果：增加训练次数提高了人工智能模型的有效性。预测人工智能计算时间的最佳拟合回归模型包括时间、样本量和培训次数变量的对数变换，其调整决定系数为 0.99：研究结果表明，通过对时间、样本量和培训课程变量进行对数变换，然后应用通过减少样本量和培训课程的若干试点研究估算出的系数，可以估算出人工智能开发所需的时间。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Orthodontics & Craniofacial Research 医学-牙科与口腔外科

CiteScore

5.30

自引率

3.20%

发文量

审稿时长

>12 weeks

期刊介绍： Orthodontics & Craniofacial Research - Genes, Growth and Development is published to serve its readers as an international forum for the presentation and critical discussion of issues pertinent to the advancement of the specialty of orthodontics and the evidence-based knowledge of craniofacial growth and development. This forum is based on scientifically supported information, but also includes minority and conflicting opinions. The objective of the journal is to facilitate effective communication between the research community and practicing clinicians. Original papers of high scientific quality that report the findings of clinical trials, clinical epidemiology, and novel therapeutic or diagnostic approaches are appropriate submissions. Similarly, we welcome papers in genetics, developmental biology, syndromology, surgery, speech and hearing, and other biomedical disciplines related to clinical orthodontics and normal and abnormal craniofacial growth and development. In addition to original and basic research, the journal publishes concise reviews, case reports of substantial value, invited essays, letters, and announcements. The journal is published quarterly. The review of submitted papers will be coordinated by the editor and members of the editorial board. It is policy to review manuscripts within 3 to 4 weeks of receipt and to publish within 3 to 6 months of acceptance.