Portal de Periódicos da CAPES

Artigo Acesso aberto Produção Nacional

BIBTEX RIS

Adaptive Crisp Active Contour Method for Segmentation and Reconstruction of 3D Lung Structures

2015; Volume: 111; Issue: 4 Linguagem: Inglês

10.5120/19523-1164

0975-8887

Pedro P. Rebouças Filho, Róger Moura Sarmento, Paulo César Cortez, Antonio Barros, Victor Hugo C. de Albuquerque,

Advanced Neural Network Applications

Computing systems have been playing an important role in various medical fields, notably in image diagnosis.Highlighted among the existing exams that allow diagnostic aids and the application of computing systems in parallel is Computed Tomography (CT).This work focuses on the segmentation and reconstruction phases of CT lung images using the Adaptive Crisp Active Contour Model 2D (ACACM) and the OpenGL library to present and analyse the results in three dimensions.The results of the proposed method were compared with those of the 3D Region Growing method and then evaluated by two pulmonologists.The results showed the superiority of the proposed method, thus confirming that that this method could integrate medical diagnostic aid systems in the pulmonology field.Finally, some applications are shown utiizando segmentation and 3D reconstruction proposals demonstrating that the proposed method can be used to aid in medical diagnosis.