Optimizing parameters of an open-source airway segmentation algorithm using different CT images

Abstract

Background Computed tomography (CT) helps physicians locate and diagnose pathological conditions. In some conditions, having an airway segmentation method which facilitates reconstruction of the airway from chest CT images can help hugely in the assessment of lung diseases. Many efforts have been made to develop airway segmentation algorithms, but methods are usually not optimized to be reliable across different CT scan parameters.MethodsIn this paper, we present a simple and reliable semi-automatic algorithm which can segment tracheal and bronchial anatomy using the open-source 3D Slicer platform. The method is based on a region growing approach where trachea, right and left bronchi are cropped and segmented independently using three different thresholds. The algorithm and its parameters have been optimized to be efficient across different CT scan acquisition parameters. The performance of the proposed method has been evaluated on EXACT’09 cases and local clinical cases as well as on a breathing pig lung phantom using multiple scans and changing parameters. In particular, to investigate multiple scan parameters reconstruction kernel, radiation dose and slice thickness have been considered. Volume, branch count, branch length and leakage presence have been evaluated. A new method for leakage evaluation has been developed and correlation between segmentation metrics and CT acquisition parameters has been considered.ResultsAll the considered cases have been segmented successfully with good results in terms of leakage presence. Results on clinical data are comparable to other teams’ methods, as obtained by evaluation against the EXACT09 challenge, whereas results obtained from the phantom prove the reliability of the method across multiple CT platforms and acquisition parameters. As expected, slice thickness is the parameter affecting the results the most, whereas reconstruction kernel and radiation dose seem not to particularly affect airway segmentation.ConclusionThe system represents the first open-source airway segmentation platform. The quantitative evaluation approach presented represents the first repeatable system evaluation tool for like-for-like comparison between different airway segmentation platforms. Results suggest that the algorithm can be considered stable across multiple CT platforms and acquisition parameters and can be considered as a starting point for the development of a complete airway segmentation algorithm.

Volumetric pulp chambers measurements in mandibular and maxillary permanent first molar using cone-beam computed tomography by age and gender

“…Yang et al used "iDixel", Porto et al used "DentalSlice", Ge et al used "ITK-SNAP 2.4", Vankatesh et al used "Mimics", and Marković et al used "3D Slicer" software as a semi-automatic program, in which CBCT images are segmented and volume measurements can be made after restructuring (15,16,21). In our study, pulpal chamber was segmented and volume measurements were made with 3D Slicer software, which was tested by Harvard Surgery Planning Laboratory and used in several studies in health sciences (34)(35)(36)(37)(38)(39)(40)(41)(42)(43).…”

Endobronchial high-intensity ultrasound for thermal therapy of pulmonary malignancies: simulations with patient-specific lung models

“…To evaluate the feasibility and performance of the endobronchial ultrasound applicators for thermal therapy of lung and bronchial cancer in a more realistic and complex anatomical setting, a total of five distinct tumor models within an anatomical lung airway model were generated, as shown in Figure 4. The lung airway structure with at least 2 mm lumen diameter of bronchi was segmented from the CT images of an adult patient with lung cancer using an airway segmentation algorithm developed by Nardelli et al [62], within the open source software platform 3D slicer (www.slicer.org). A total of five tumors between 1 and 3 cm diameter were segmented using iSeg (Zurich MedTech AG, Zurich, Switzerland) and were placed at different sites to represent a range of patient cases.…”

Three-dimensional Printing and 3D Slicer

“…One of the main advantages of 3D Slicer is that it provides an extensible platform for novel automated algorithms. In a recent publication by Nardelli et al, 37 an airway segmentation algorithm was developed as a 3D Slicer extension and validated with a number of CT scans. Most airway segmentation techniques are based on a region-growing approach in which a base voxel serves as the seed and a threshold cutoff for separating air from tissue.…”

“…Most airway segmentation techniques are based on a region-growing approach in which a base voxel serves as the seed and a threshold cutoff for separating air from tissue. The method of Nardelli et al 37 used region growing from cropped trachea and right and left bronchi with their respective thresholds. This approach allowed them to generate a very accurate tracheobronchial tree (Video 4; e-Appendix 1).…”