Abstract:HighlightsThis manuscript proposes a novel recursive strategy based on geometric active contour model to extract lung tissues from the volumetric CT slices accurately.The proposed method could settle the challenging task of separating left and right lungs.Non-sheltered 3D models of lung and vessels tree are constructed based on the extracted datasets.The proposed method is validated by fifteen scans with good performance. Three dimensional reconstruction of lung and vessel tree has great significance to 3D observation and quantitative analysis for lung diseases. This paper presents non-sheltered 3D models of lung and vessel tree based on a supervised semi-3D lung tissues segmentation method. A recursive strategy based on geometric active contour is proposed instead of the "coarse-to-fine" framework in existing literature to extract lung tissues from the volumetric CT slices. In this model, the segmentation of the current slice is supervised by the result of the previous one slice due to the slight changes between adjacent slice of lung tissues. Through this mechanism, lung tissues in all the slices are segmented fast and accurately. The serious problems of left and right lungs fusion, caused by partial volume effects, and segmentation of pleural nodules can be settled meanwhile during the semi-3D process. The proposed scheme is evaluated by fifteen scans, from eight healthy participants and seven participants suffering from early-stage lung tumors. The results validate the good performance of the proposed method compared with the "coarse-to-fine" framework. The segmented datasets are utilized to reconstruct the non-sheltered 3D models of lung and vessel tree.

Virtual Endoscopy in Branch Structure

Research and Realization of Virtual Endoscopy

Research on Virtual Endoscopy System.

Slice-based Virtual Endoscopy Navigation

Research on key technologies of virtual endoscopy

Realization technologies in a virtual endoscopy system

Navigation and Intervention of Three-Dimensional Echocardiographic Virtual Endoscopic System: a Primary Methodological Study

Airway segmentation from 3D chest CT volumes based on volume of interest using gradient vector flow

Constrained tetrahedral mesh generation of human organs on segmented volume

A New Technique of the Virtual Reality Visualization of Complex Volume Images from the Computer Tomography and Magnetic Resonance Imaging

Visualization Simulation of Branch Fractures Based on Internal Structure Reconstruction

Supervised recursive segmentation of volumetric CT images for 3D reconstruction of lung and vessel tree

Virtual Measurement in Reverse CAD Modeling of Organ

New Fast Algorithm for Extracting Center Path

Immersive Anatomical Scenes that Enable Multiple Users to Occupy the Same Virtual Space: A Tool for Surgical Planning and Education

Accelerating the Volume Rendering in Virtual Endoscopy

Analytical Solutions for Tree‐like Structure Modelling Using Subdivision Surfaces

Research of the Key Technique in Virtual Endoscopy and Its Clinical Application

BREAK: BRONCHI RECONSTRUCTION BY GEODESIC TRANSFORMATION AND SKELETON EMBEDDING

Application of Virtual Reality in Quantified Analyses for Cavernous Sinus Surgical Approach

Study on Visualization of Congenital Heart Disease by Three-Dimensional Echocardiographic Virtual Endoscopy