Karine Evangelista1,2, Brunno Santos de Freitas Silva3, Fernanda Paula Yamamoto-Silva3, José Valladares-Neto4, Maria Alves Garcia Silva3, Lucia Helena Soares Cevidanes5, Graziela de Luca Canto6, Carla Massignan7. 1. School of Dentistry, Federal University of Goiás, Avenida Universitária esquina com 1a Avenida, Goiânia, S/N. Zip Code: 74605-220, Brazil. kemar_7@hotmail.com. 2. Division of Orthodontics, School of Dentistry, Federal University of Goiás, Avenida Universitária esquina com 1a Avenida, Goiânia, S/N. Zip Code: 74605-220, Brazil. kemar_7@hotmail.com. 3. Department of Stomatology, School of Dentistry, Federal University of Goiás, Avenida Universitária esquina com 1a Avenida, Goiânia, S/N. Zip Code: 74605-220, Brazil. 4. Division of Orthodontics, School of Dentistry, Federal University of Goiás, Avenida Universitária esquina com 1a Avenida, Goiânia, S/N. Zip Code: 74605-220, Brazil. 5. Department of Orthodontics and Pediatric Dentistry, School of Dentistry, University of Michigan, 1011 N University Ave, Ann Arbor, MI, Zip Code: 48109, USA. 6. Department of Dentistry, Brazilian Centre for Evidence-Based Research, Health Sciences Center, Federal University of Santa Catarina, Rua Delfino Conti, 1240-Trindade, Florianópolis, Zip Code: 88040-535, Brazil. 7. Department of Dentistry, University of Brasilia, UnB Estac. Medicina UnB-Asa Norte, Brasilia, Zip Code: 70297-400, Brazil.
Abstract
OBJECTIVE: This study aimed to analyze the accuracy of artificial intelligence (AI) for orthodontic tooth extraction decision-making. MATERIALS AND METHODS: PubMed/MEDLINE, EMBASE, LILACS, Web of Science, Scopus, LIVIVO, Computers & Applied Science, ACM Digital Library, Compendex, and gray literature (OpenGrey, ProQuest, and Google Scholar) were electronically searched. Three independent reviewers selected the studies and extracted and analyzed the data. Risk of bias, methodological quality, and certainty of evidence were assessed by QUADAS-2, checklist for AI research, and GRADE, respectively. RESULTS: The search identified 1810 studies. After 2 phases of selection, six studies were included, showing an unclear risk of bias of patient selection. Two studies showed a high risk of bias in the index test, while two others presented an unclear risk of bias in the diagnostic test. Data were pooled in a random model and yielded an accuracy value of 0.87 (95% CI = 0.75-0.96) for all studies, 0.89 (95% CI = 0.70-1.00) for multilayer perceptron, and 0.88 (95% CI = 0.73-0.98) for back propagation models. Sensitivity, specificity, and area under the curve of the multilayer perceptron model yielded 0.84 (95% CI = 0.58-1.00), 0.89 (95% CI = 0.74-0.98), and 0.92 (95% CI = 0.72-1.00) scores, respectively. Sagittal discrepancy, upper crowding, and protrusion showed the highest ranks weighted in the models. CONCLUSIONS: Orthodontic tooth extraction decision-making using AI presented promising accuracy but should be considered with caution due to the very low certainty of evidence. CLINICAL RELEVANCE: AI models for tooth extraction decision in orthodontics cannot yet be considered a substitute for a final human decision.
OBJECTIVE: This study aimed to analyze the accuracy of artificial intelligence (AI) for orthodontic tooth extraction decision-making. MATERIALS AND METHODS: PubMed/MEDLINE, EMBASE, LILACS, Web of Science, Scopus, LIVIVO, Computers & Applied Science, ACM Digital Library, Compendex, and gray literature (OpenGrey, ProQuest, and Google Scholar) were electronically searched. Three independent reviewers selected the studies and extracted and analyzed the data. Risk of bias, methodological quality, and certainty of evidence were assessed by QUADAS-2, checklist for AI research, and GRADE, respectively. RESULTS: The search identified 1810 studies. After 2 phases of selection, six studies were included, showing an unclear risk of bias of patient selection. Two studies showed a high risk of bias in the index test, while two others presented an unclear risk of bias in the diagnostic test. Data were pooled in a random model and yielded an accuracy value of 0.87 (95% CI = 0.75-0.96) for all studies, 0.89 (95% CI = 0.70-1.00) for multilayer perceptron, and 0.88 (95% CI = 0.73-0.98) for back propagation models. Sensitivity, specificity, and area under the curve of the multilayer perceptron model yielded 0.84 (95% CI = 0.58-1.00), 0.89 (95% CI = 0.74-0.98), and 0.92 (95% CI = 0.72-1.00) scores, respectively. Sagittal discrepancy, upper crowding, and protrusion showed the highest ranks weighted in the models. CONCLUSIONS: Orthodontic tooth extraction decision-making using AI presented promising accuracy but should be considered with caution due to the very low certainty of evidence. CLINICAL RELEVANCE: AI models for tooth extraction decision in orthodontics cannot yet be considered a substitute for a final human decision.
Authors: Tate H Jackson; Camille Guez; Feng-Chang Lin; William R Proffit; Ching-Chang Ko Journal: Am J Orthod Dentofacial Orthop Date: 2017-03 Impact factor: 2.650