Prediction of the relationship between the mandibular and craniofacial bone and cervical vertebrae based on a genetic algorithm in patients with skeletal class Ⅱ malocclusion

To explore the correlation between the parameters of the mandible and parameters of cervical vertebrae and craniofacial bone in class Ⅱ skeletal patients in Northeast China and to establish correlation equations expressing the relationship between the mandible and cervical vertebrae and craniofacial bone directly and quantitatively for the clinical diagnosis and treatment of orthodontics and orthognathics and for prediction.

The mandible, cranial facial bone and cervical vertebrae of 201 children and adolescents aged 8 to 20 years were measured using digital cranial lateral tablets. All of the cases were divided into male (n = 75) and female (n = 126) groups using a sensitivity analysis method based on genetic algorithms to select the craniofacial bone and cervical bone with strong sensitivity to mandible parameters and to establish relevant equations.

Through sensitivity analysis, the parameters with the strongest correlation between the measured values of the mandible were H4 and SN, those with a strong correlation were SN-Ar, the anterior and posterior high ratio SGo/NGn, the Y axis angle and mandibular angle Ar-Go-Gn. The established equation was as follows: males: Ar-Pg=28.415+1.818×H4+0.746×SN(r² = 0.056 8, P < 0.001); females: Ar-Pg=15.168+1.706×H4+0.675×SN+0.31×SN-Ar-0.29×Y axis angle (r² = 0.611, P < 0.001). No significant difference was found between the predicted values obtained by the established equations and measured values (P > 0.05).

The mandibular length equation established by sensitivity analysis and genetic algorithms is statistically significant and can predict a certain degree of growth and development.