{"title":"Comparative analysis of facial aesthetics in AI generated versus conventionally crafted digital smile designs-a cross-sectional study.","authors":"Kriti Kaushik, Ann Sales, Shobha J Rodrigues","doi":"10.1038/s41405-025-00367-z","DOIUrl":null,"url":null,"abstract":"<p><strong>Aim: </strong>This study aimed to evaluate the aesthetic preferences of traditional digital smile designs and artificial intelligence (AI)-generated smile designs among dentists, dental students, and laypersons, addressing gaps in previous research on the clinical acceptability of AI in prosthodontic aesthetics.</p><p><strong>Materials and methods: </strong>A cross-sectional, questionnaire-based study was conducted via an online survey distributed across India between 2024 and 2025. A total of 320 participants, including dental students, dentists, and nondental professionals, were recruited on the basis of calculated sample size requirements. Smile designs were created for four clinical cases via Exo-CAD software, employing two methods: conventional manual design by prosthodontists and AI-based automated design. The participants evaluated paired smile designs and indicated their aesthetic preferences. Demographic data were also collected. Chi-square (χ²) tests were applied for statistical analysis, with a significance level set at p < 0.05.</p><p><strong>Results: </strong>No significant differences in aesthetic preferences were observed based on sex, age, or occupation. Overall, manually crafted smile designs were consistently preferred across all the participant categories. However, AI-generated smiles for Cases 3 and 4 presented relatively higher acceptance rates (39.4% and 39.7%, respectively) than those for Cases 1 and 2 did. The findings suggest that while AI algorithms can achieve acceptable levels of aesthetic appeal, they still lack the human touch essential for capturing nuanced facial dynamics and emotional context.</p><p><strong>Conclusion: </strong>Although AI-based smile design systems demonstrate promise in improving workflow efficiency and consistency, they are currently unable to replicate the individualized artistic judgment of experienced clinicians. Manual intervention remains critical for achieving truly personalized and aesthetically harmonious outcomes. Future approaches should consider hybrid models that combine AI automation with clinician-led customization to increase both the efficiency and patient satisfaction of smile aesthetics.</p>","PeriodicalId":36997,"journal":{"name":"BDJ Open","volume":"11 1","pages":"79"},"PeriodicalIF":2.5000,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12436661/pdf/","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"BDJ Open","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1038/s41405-025-00367-z","RegionNum":0,"RegionCategory":null,"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
Aim: This study aimed to evaluate the aesthetic preferences of traditional digital smile designs and artificial intelligence (AI)-generated smile designs among dentists, dental students, and laypersons, addressing gaps in previous research on the clinical acceptability of AI in prosthodontic aesthetics.
Materials and methods: A cross-sectional, questionnaire-based study was conducted via an online survey distributed across India between 2024 and 2025. A total of 320 participants, including dental students, dentists, and nondental professionals, were recruited on the basis of calculated sample size requirements. Smile designs were created for four clinical cases via Exo-CAD software, employing two methods: conventional manual design by prosthodontists and AI-based automated design. The participants evaluated paired smile designs and indicated their aesthetic preferences. Demographic data were also collected. Chi-square (χ²) tests were applied for statistical analysis, with a significance level set at p < 0.05.
Results: No significant differences in aesthetic preferences were observed based on sex, age, or occupation. Overall, manually crafted smile designs were consistently preferred across all the participant categories. However, AI-generated smiles for Cases 3 and 4 presented relatively higher acceptance rates (39.4% and 39.7%, respectively) than those for Cases 1 and 2 did. The findings suggest that while AI algorithms can achieve acceptable levels of aesthetic appeal, they still lack the human touch essential for capturing nuanced facial dynamics and emotional context.
Conclusion: Although AI-based smile design systems demonstrate promise in improving workflow efficiency and consistency, they are currently unable to replicate the individualized artistic judgment of experienced clinicians. Manual intervention remains critical for achieving truly personalized and aesthetically harmonious outcomes. Future approaches should consider hybrid models that combine AI automation with clinician-led customization to increase both the efficiency and patient satisfaction of smile aesthetics.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
