A method for avoiding overlap of left and right lungs in shape model guided segmentation of lungs in CT volumes

• Published in: Medical physics (Lancaster) 2014-10, Vol.41 (10), p.101908-n/a
• Main Authors: Gill, Gurman, Bauer, Christian, Beichel, Reinhard R.
• Format: Article
• Language: English
• Subjects: active shape model; Algorithms; Asthma - diagnostic imaging; Biological material, e.g. blood, urine; Haemocytometers; biological tissues; Computed tomography; Computerised tomographs; computerised tomography; Digital computing or data processing equipment or methods, specially adapted for specific applications; diseases; Eigenvalues; Gold; Humans; image classification; Image data processing or generation, in general; image segmentation; lung; Lung - diagnostic imaging; lung segmentation; lung separation; Lungs; medical image processing; Medical image segmentation; Models, Biological; Pattern Recognition, Automated - methods; Pulmonary Disease, Chronic Obstructive - diagnostic imaging; Radiation Imaging Physics; Respiration; Segmentation; Three dimensional image processing; Three dimensional sensing; Tissues; Tomography, X-Ray Computed - methods
• ISSN: 0094-2405; 2473-4209
• DOI: 10.1118/1.4894817

Purpose: The automated correct segmentation of left and right lungs is a nontrivial problem, because the tissue layer between both lungs can be quite thin. In the case of lung segmentation with left and right lung models, overlapping segmentations can occur. In this paper, the authors address this issue and propose a solution for a model‐based lung segmentation method. Methods: The thin tissue layer between left and right lungs is detected by means of a classification approach and utilized to selectively modify the cost function of the lung segmentation method. The approach was evaluated on a diverse set of 212 CT scans of normal and diseased lungs. Performance was assessed by utilizing an independent reference standard and by means of comparison to the standard segmentation method without overlap avoidance. Results: For cases where the standard approach produced overlapping segmentations, the proposed method significantly (p = 1.65 × 10−9) reduced the overlap by 97.13% on average (median: 99.96%). In addition, segmentation accuracy assessed with the Dice coefficient showed a statistically significant improvement (p = 7.5 × 10−5) and was 0.9845 ± 0.0111. For cases where the standard approach did not produce an overlap, performance of the proposed method was not found to be significantly different. Conclusions: The proposed method improves the quality of the lung segmentations, which is important for subsequent quantitative analysis steps.