Secção anatômica manual e automatizada de modelos 3D de côndilos mandibulares: proposição de método e avaliação de confiabilidade

• Autor(a) principal: Carneiro, Mario Sergio Alves
• Data de Publicação: 2016
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Biblioteca Digital de Teses e Dissertações da UERJ
• Texto Completo: http://www.bdtd.uerj.br/handle/1/14061
• topic: Cone-Beam Computed Tomography; Imaging Three-Dimensional; Mandibular Condyle; Computer Simulation; Data Accuracy; Tomografia Computadorizada de Feixe Cônico; Imagem Tridimensional; Côndilo Mandibular; Simulação por Computador; Confiabilidade dos Dados; CNPQ::CIENCIAS DA SAUDE::ODONTOLOGIA::PERIODONTIA

This study aimed to determine the sectional planes of the mandibular condyles and evaluate the reliability of two tridimensional images (3D) sections methods: manual and automated. The 3D images were acquired using computed tomography cone-beam (CBCT). Eighty one CBCTs were selected from 27 patients from database of patients undergoing isolated mandibular advancement surgery in the University of North Carolina - UNC Memorial Hospital). CBCTs were obtained for each individual in three different times: before surgery (T1); after surgical removal of the eaves - 4 to 6 weeks post-surgery (T2) and one year after surgery (T3). Virtual models (3D) of the condyles of all individuals in these three times were built. After superposition of the models was performed in T1, T2 and T3, 9 images were randomly selected, from 8 patients (a patient had both condyles selected) and the section of them was performed by two evaluators using both methods: manual and automated. The condyle was divided into 4 anatomic sections: anterolateral (AL), anteromedial (AM), posterolateral (PL) and posteromedial (PM). Both methods were repeated with 2 week intervals to assess the reproducibility by 2 reviewers.The Wilcoxon test was used to verify if the recorded medians for each comparison were statistically different from 0, as well as, to test if it were significant intra-raters differences among the evaluated methods. For the inter-raters comparison between the different methods the Mann-Whitney U test was used. The results showed that all the evaluated regions showed significant differences when compared to 0 for intra and inter-raters comparisons for both methods (manual and authomated) (p < 0,05). The inter-raters comparison revealed statistically significant differences for both methods in all the evaluated regions. The intra-raters comparison showed statistically significant differences for the following regions: AM, PL e PM (rater 1) and only AL (rater 2) (p < 0,05). When considering a tolerance of 1,5 mm, the intra-rater agreement for the manual method was (rater 1=74% and rater 2=94,5%) and for the authomated method (rater 1=89% and rater 2=96,5%). Considering the upcited tolerance the recorded inter-raters agreement was 49% for the manual method and 78% for the authomated method. It was concluded that there the automated method were more reliable in the determination sectional planes of the mandibular condyles.