비공 모양 분석을 이용한 구순열 비변형의 객관적 평가

Abstract

Cleft lip is one of the most common congenital deformities in craniofacial region. However, despite many reports on the outcome of various surgical techniques from individual medical centers, the evaluation of the outcome has been based on the subjective observation. Therefore, developing a new technique of objective and scientific evaluation for the nasal deformity of secondary cleft lip and nasal deformity is important to improve the management of the cleft lip patients.The authors have proposed a new scoring model to evaluate objectively the nasal deformity. The authors put ’x’ axis on the nostril sill of noncleft side and ’y’ axis on nasal tip at a right angle to x axis. The nostril of noncleft side was turned around the y axis to form the mirror image on the cleft side, and following four parameters were recorded: (1) angle difference between two nostril axis, (2) center of the nostril and distance between two centers, (3) the overlapped area of two nostrils and (4) the overlapped area ratio of two nostrils. The post-surgery images were evaluated by the three plastic surgeons, using the scale of 5 percentiles. The four parameters of each image were measured using a newly developed software and correlation coefficients of each parameter and the evaluation scale by the surgeons were obtained. Among the scores from scoring model, errors greater than 10 percents were found only in 2 of 14 cases (2/14=14%), but among the scores from examiners, errors of greater than 10 percents were 4 from 14 cases (4/14=29%). Based on the results, deviation among the individual examiners can be reduced using the objective evaluation program developed in this study. The correlation coefficients between the normalized overlap area (0.78) and the distance between the centers of two nostrils (-0.82) with the evaluated scale by surgeons were high. However, the relative area and the degree difference of two nostril axis were found to be not such a good parameters in the evaluation of the nasal deformity. We developed scoring model from analysis of nostril shape and neural network which is able to evaluate cleft lip nasal deformity objectively after selection of proper parameters.