Two-dimensional airway analysis using probabilistic neural networks

Jun Tan, Bin Zheng, Sang Cheol Park, Jiantao Pu, Frank C. Sciurba, Joseph K. Leader

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Although 3-D airway tree segmentation permits analysis of airway tree paths of practical lengths and facilitates visual inspection, our group developed and tested an automated computer scheme that was operated on individual 2-D CT images to detect airway sections and measure their morphometry and/or dimensions. The algorithm computes a set of airway features including airway lumen area (Ai), airway cross-sectional area (Aw), the ratio (Ra) of Ai to Aw, and the airway wall thickness (Tw) for each detected airway section depicted on the CT image slice. Thus, this 2-D based algorithm does not depend on the accuracy of 3-D airway tree segmentation and does not require that CT examination encompasses the entire lung or reconstructs contiguous images. However, one disadvantage of the 2-D image based schemes is the lack of the ability to identify the airway generation (G b) of the detected airway section. In this study, we developed and tested a new approach that uses 2-D airway features to assign a generation number to an airway. We developed and tested two probabilistic neural networks (PNN) based on different sets of airway features computed by our 2-D based scheme. The PNNs were trained and tested on 12 lung CT examinations (8 training and 4 testing). The accuracy for the PNN that utilized Ai and Ra for identifying the generation of airway sections varies from 55.4% - 100%. The overall accuracy of the PNN for all detected airway sections that are spread over all generations is 76.7%. Interestingly, adding wall thickness feature (T w) to PNN did not improve identification accuracy. This preliminary study demonstrates that a set of 2-D airway features may be used to identify the generation number of an airway with reasonable accuracy.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging
Volume7626
DOIs
StatePublished - Jun 15 2010
EventMedical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging - San Diego, CA, United States
Duration: Feb 14 2010Feb 16 2010

Other

OtherMedical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging
CountryUnited States
CitySan Diego, CA
Period2/14/102/16/10

Fingerprint

dimensional analysis
Neural networks
Lung
lungs
examination
lumens
inspection
education

Keywords

  • Airway detection
  • Airway generation
  • Computer-Aided Detection and Diagnosis (CAD)
  • Probabilistic Neural Network (PNN)

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Atomic and Molecular Physics, and Optics
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Cite this

Tan, J., Zheng, B., Park, S. C., Pu, J., Sciurba, F. C., & Leader, J. K. (2010). Two-dimensional airway analysis using probabilistic neural networks. In Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging (Vol. 7626). [762612] https://doi.org/10.1117/12.844497

Two-dimensional airway analysis using probabilistic neural networks. / Tan, Jun; Zheng, Bin; Park, Sang Cheol; Pu, Jiantao; Sciurba, Frank C.; Leader, Joseph K.

Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging. Vol. 7626 2010. 762612.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Tan, J, Zheng, B, Park, SC, Pu, J, Sciurba, FC & Leader, JK 2010, Two-dimensional airway analysis using probabilistic neural networks. in Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging. vol. 7626, 762612, Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging, San Diego, CA, United States, 2/14/10. https://doi.org/10.1117/12.844497
Tan J, Zheng B, Park SC, Pu J, Sciurba FC, Leader JK. Two-dimensional airway analysis using probabilistic neural networks. In Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging. Vol. 7626. 2010. 762612 https://doi.org/10.1117/12.844497
Tan, Jun ; Zheng, Bin ; Park, Sang Cheol ; Pu, Jiantao ; Sciurba, Frank C. ; Leader, Joseph K. / Two-dimensional airway analysis using probabilistic neural networks. Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging. Vol. 7626 2010.
@inproceedings{34287d4c0aec4383ab29a5428e49994a,
title = "Two-dimensional airway analysis using probabilistic neural networks",
abstract = "Although 3-D airway tree segmentation permits analysis of airway tree paths of practical lengths and facilitates visual inspection, our group developed and tested an automated computer scheme that was operated on individual 2-D CT images to detect airway sections and measure their morphometry and/or dimensions. The algorithm computes a set of airway features including airway lumen area (Ai), airway cross-sectional area (Aw), the ratio (Ra) of Ai to Aw, and the airway wall thickness (Tw) for each detected airway section depicted on the CT image slice. Thus, this 2-D based algorithm does not depend on the accuracy of 3-D airway tree segmentation and does not require that CT examination encompasses the entire lung or reconstructs contiguous images. However, one disadvantage of the 2-D image based schemes is the lack of the ability to identify the airway generation (G b) of the detected airway section. In this study, we developed and tested a new approach that uses 2-D airway features to assign a generation number to an airway. We developed and tested two probabilistic neural networks (PNN) based on different sets of airway features computed by our 2-D based scheme. The PNNs were trained and tested on 12 lung CT examinations (8 training and 4 testing). The accuracy for the PNN that utilized Ai and Ra for identifying the generation of airway sections varies from 55.4{\%} - 100{\%}. The overall accuracy of the PNN for all detected airway sections that are spread over all generations is 76.7{\%}. Interestingly, adding wall thickness feature (T w) to PNN did not improve identification accuracy. This preliminary study demonstrates that a set of 2-D airway features may be used to identify the generation number of an airway with reasonable accuracy.",
keywords = "Airway detection, Airway generation, Computer-Aided Detection and Diagnosis (CAD), Probabilistic Neural Network (PNN)",
author = "Jun Tan and Bin Zheng and Park, {Sang Cheol} and Jiantao Pu and Sciurba, {Frank C.} and Leader, {Joseph K.}",
year = "2010",
month = "6",
day = "15",
doi = "10.1117/12.844497",
language = "English (US)",
isbn = "9780819480279",
volume = "7626",
booktitle = "Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging",

}

TY - GEN

T1 - Two-dimensional airway analysis using probabilistic neural networks

AU - Tan, Jun

AU - Zheng, Bin

AU - Park, Sang Cheol

AU - Pu, Jiantao

AU - Sciurba, Frank C.

AU - Leader, Joseph K.

PY - 2010/6/15

Y1 - 2010/6/15

N2 - Although 3-D airway tree segmentation permits analysis of airway tree paths of practical lengths and facilitates visual inspection, our group developed and tested an automated computer scheme that was operated on individual 2-D CT images to detect airway sections and measure their morphometry and/or dimensions. The algorithm computes a set of airway features including airway lumen area (Ai), airway cross-sectional area (Aw), the ratio (Ra) of Ai to Aw, and the airway wall thickness (Tw) for each detected airway section depicted on the CT image slice. Thus, this 2-D based algorithm does not depend on the accuracy of 3-D airway tree segmentation and does not require that CT examination encompasses the entire lung or reconstructs contiguous images. However, one disadvantage of the 2-D image based schemes is the lack of the ability to identify the airway generation (G b) of the detected airway section. In this study, we developed and tested a new approach that uses 2-D airway features to assign a generation number to an airway. We developed and tested two probabilistic neural networks (PNN) based on different sets of airway features computed by our 2-D based scheme. The PNNs were trained and tested on 12 lung CT examinations (8 training and 4 testing). The accuracy for the PNN that utilized Ai and Ra for identifying the generation of airway sections varies from 55.4% - 100%. The overall accuracy of the PNN for all detected airway sections that are spread over all generations is 76.7%. Interestingly, adding wall thickness feature (T w) to PNN did not improve identification accuracy. This preliminary study demonstrates that a set of 2-D airway features may be used to identify the generation number of an airway with reasonable accuracy.

AB - Although 3-D airway tree segmentation permits analysis of airway tree paths of practical lengths and facilitates visual inspection, our group developed and tested an automated computer scheme that was operated on individual 2-D CT images to detect airway sections and measure their morphometry and/or dimensions. The algorithm computes a set of airway features including airway lumen area (Ai), airway cross-sectional area (Aw), the ratio (Ra) of Ai to Aw, and the airway wall thickness (Tw) for each detected airway section depicted on the CT image slice. Thus, this 2-D based algorithm does not depend on the accuracy of 3-D airway tree segmentation and does not require that CT examination encompasses the entire lung or reconstructs contiguous images. However, one disadvantage of the 2-D image based schemes is the lack of the ability to identify the airway generation (G b) of the detected airway section. In this study, we developed and tested a new approach that uses 2-D airway features to assign a generation number to an airway. We developed and tested two probabilistic neural networks (PNN) based on different sets of airway features computed by our 2-D based scheme. The PNNs were trained and tested on 12 lung CT examinations (8 training and 4 testing). The accuracy for the PNN that utilized Ai and Ra for identifying the generation of airway sections varies from 55.4% - 100%. The overall accuracy of the PNN for all detected airway sections that are spread over all generations is 76.7%. Interestingly, adding wall thickness feature (T w) to PNN did not improve identification accuracy. This preliminary study demonstrates that a set of 2-D airway features may be used to identify the generation number of an airway with reasonable accuracy.

KW - Airway detection

KW - Airway generation

KW - Computer-Aided Detection and Diagnosis (CAD)

KW - Probabilistic Neural Network (PNN)

UR - http://www.scopus.com/inward/record.url?scp=77953302061&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77953302061&partnerID=8YFLogxK

U2 - 10.1117/12.844497

DO - 10.1117/12.844497

M3 - Conference contribution

AN - SCOPUS:77953302061

SN - 9780819480279

VL - 7626

BT - Medical Imaging 2010 - Biomedical Applications in Molecular, Structural, and Functional Imaging

ER -