Design and implementation of a hybrid genetic algorithm and artificial neural network system for predicting the sizes of unerupted canines and premolars

• Published in: European journal of orthodontics Vol. 34; no. 4; pp. 480 - 486
• Main Authors: Moghimi, S., Talebi, M., Parisay, I.
• Format: Journal Article
• Language: English
• Published: England 01.08.2012
• Subjects: Adolescent; Algorithms; Bicuspid - anatomy & histology; Cuspid - anatomy & histology; Dentition, Mixed; Female; Humans; Male; Models, Genetic; Neural Networks (Computer); Odontometry - methods; Odontometry - statistics & numerical data; Reference Standards; Regression Analysis; Tooth Crown - anatomy & histology; Tooth, Unerupted - anatomy & histology
• ISSN: 0141-5387; 1460-2210; 1460-2210
• DOI: 10.1093/ejo/cjr042

The aim of this study was to develop a novel hybrid genetic algorithm and artificial neural network (GA-ANN) system for predicting the sizes of unerupted canines and premolars during the mixed dentition period. This study was performed on 106 untreated subjects (52 girls, 54 boys, aged 13-15 years). Data were obtained from dental cast measurements. A hybrid GA-ANN algorithm was developed to find the best reference teeth and the most accurate mapping function. Based on a regression analysis, the strongest correlation was observed between the sum of the mesiodistal widths of the mandibular canines and premolars and the mesiodistal widths of the mandibular first molars and incisors (r = 0.697). In the maxilla, the highest correlation was observed between the sum of the mesiodistal widths of the canines and premolars and the mesiodistal widths of the mandibular first molars and maxillary central incisors (0.742). The hybrid GA-ANN algorithm selected the mandibular first molars and incisors and the maxillary central incisors as the reference teeth for predicting the sum of the mesiodistal widths of the canines and premolars. The prediction error rates and maximum rates of over/underestimation using the hybrid GA-ANN algorithm were smaller than those using linear regression analyses.