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

Authors

  • Gill Gurman,

    1. Department of Electrical and Computer Engineering, The University of Iowa, Iowa City, Iowa 52242 and The Iowa Institute for Biomedical Imaging, The University of Iowa, Iowa City, Iowa 52242
    Search for more papers by this author
  • Bauer Christian,

    1. Department of Electrical and Computer Engineering, The University of Iowa, Iowa City, Iowa 52242 and The Iowa Institute for Biomedical Imaging, The University of Iowa, Iowa City, Iowa 52242
    Search for more papers by this author
  • Beichel Reinhard R.

    Corresponding author
    1. Department of Electrical and Computer Engineering, The University of Iowa, Iowa City, Iowa 52242; The Iowa Institute for Biomedical Imaging, The University of Iowa, Iowa City, Iowa 52242; and Department of Internal Medicine, The University of Iowa, Iowa City, Iowa 52242
    Search for more papers by this author

Abstract

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.

Ancillary