3D reconstruction of pulmonary nodules in PET-CT image sequences based on a novel 3D region growing method combined with ACO. (2018)

• Record Type: Journal Article
• Authors: Zhao, Juan-juan; Qiang, Wei; Ji, Guo-hua; Zhou, Xiang-fei
• Is Part Of: International journal of bio-inspired computation. Volume 11:Number 1(2018)
• Page Start: 54
• Page End: 59
• Publication Date: 2018
• Subjects: pulmonary nodules -- 3D visualisation -- k-means -- 3D region growing -- ant colony optimisation -- ACO; Biologically-inspired computing -- Periodicals; Computational biology -- Periodicals; 572.0285
• Languages: English
• ISSNs: 1758-0366

Abstract

The three-dimensional visualisation is an important aid for the detection and diagnosis of pulmonary nodules. The traditional method by which clinicians restore the 3D structure of pulmonary nodules (i.e., by subjective imagination and clinical experience, which may not be intuitive or accurate) is not conducive to pulmonary nodule extraction and quantification. Therefore, we herein propose an algorithm of pulmonary nodule segmentation and 3D reconstruction based on 3D region growing in positron emission tomography-computed tomography (PET-CT) image sequences. First, k-means clustering was used for the lung parenchyma segmentation. Next, 3D surface rendering reconstruction of lung parenchyma was performed. Finally, the novel 3D region growing method optimised by ant colony optimisation (ACO) was used to segment the pulmonary nodule. Our proposed method was more efficient than traditional methods in the present study. The experimental results show that our algorithm can segment pulmonary nodules more fully with high segmentation precision and accuracy.