Extraction of Airways From CT (EXACT'09)

Abstract

Abstract. This paper describes a framework for evaluating airway extraction algorithms in a standardized manner and establishing reference segmentations that can be used for future algorithm development. Because of the sheer difficulty of constructing a complete reference standard manually, we propose to construct a reference using results from the algorithms being compared, by splitting each airway tree segmentation result into individual branch segments that are subsequently visually inspected by trained observers. Using the so constructed reference, a total of seven performance measures covering different aspects of segmentation quality are computed. We evaluated 15 airway tree extraction algorithms from different research groups on a diverse set of 20 chest CT scans from subjects ranging from healthy volunteers to patients with severe lung disease, who were scanned at different sites, with several different CT scanner models, and using a variety of scanning protocols and reconstruction parameters.

Deep Learning for Medical Image Analysis: Applications to Computed Tomography and Magnetic Resonance Imaging

“…Airway segmentation can also increase the accuracy of the segmentation of other structures within the thorax and lower the false positive rate in lung nodule detection. Though there have been many proposals for airway segmentation, most studies have focused on achieving high sensitivity of detecting airways, which led to a high false positive rate [19]. To tackle this problem, Charbonnier et al trained a CNN to classify leak candidates into airways and leaks from multiple initial airway segmentation results by changing the parameters of the segmentation algorithm [20].…”

Computer‐aided detection and quantification of cavitary tuberculosis from CT scans

“…As it can be simply seen from the figure, the boundary of the cavity and nearby airway structures were identified successfully. Furthermore, we tested the proposed airway segmentation algorithm quantitatively on publicly available human CT scans (EXACT09 challenge 26 ) and obtained promising results. Based on the evaluation metric provided by the challenge organizers, 26 we obtained a second best detection rate with a low false positive rate (<1%).…”

“…Furthermore, we tested the proposed airway segmentation algorithm quantitatively on publicly available human CT scans (EXACT09 challenge 26 ) and obtained promising results. Based on the evaluation metric provided by the challenge organizers, 26 we obtained a second best detection rate with a low false positive rate (<1%). Extended evaluation metrics of the segmentation challenge and results of the dataset from human CT scans is outside the scope and aim of this paper.…”

Dissimilarity-Based Classification of Anatomical Tree Structures

“…Figure 1(a) shows an example of a segmented airway tree using this algorithm. The segmentation is subdivided into branches and branch centerlines are extracted using the algorithm described in [17]. Figure 1(b) shows a coloring of the identified branches using this algorithm.…”

“…Branch generations are obtained by assigning generation 0 to the trachea and incrementing generation number by one when propagating the generation number from a parent centerline to its child centerlines. All the steps taken are fully automatic, and the reader is referred to [16,17] for further details. Each branch is represented by a 5-dimensional feature vector x i comprising one measure known to be related to COPD as well as four anatomical features roughly capturing the location and orientation of the branch in the airway tree.…”