Pulmonary diseases rank third among the leading causes of death in the United States. The availability of high resolution MDCT makes it possible to image in-vivo lung structures in previously unattainable details, thus stimulating great interest of using imaging to phenotype lung abnormalities for better understanding of lung anatomy and its correlation to lung function. This may lead to earlier detection and diagnosis as well as possibly to more efficacious individualized patient management. There are indications that fissures integrity and configuration may be correlated with lung function and may play a role in the susceptibility of individuals to the development of specific lung diseases such as COPD (Chronic Obstructive Pulmonary Disease) and ILD (Interstitial Lung Disease). Fissure integrity may also be a primary factor in determining progression of diseases such as pneumonia. However, subjective assessments of fissures completeness and configuration as well as surrounding tissues are typically done subjectively by radiologists and pulmonologists and these assessments are both time consuming and error prone. To take advantages of available advanced CT imaging techniques and to provide a non-invasive way for assessing these relationships, we propose to develop automated computerized schemes to visualize and quantitatively measure and classify pulmonary fissures structure and completeness and then assess correlations, if any, between established lung function measures and differences in lung structures. To achieve these objectives, we will (1) assemble a large and diverse CT database of over 1000 lung CT examinations that cover a wide range of depicted lung diseases, (2) optimize a previously developed prototype computerized scheme for lung fissure detection and lobe segmentation and test its generalizeability, (3) develop an automated quantitative computerized tool for computing a summary index for fissure configuration and fissure incompleteness, (4) evaluate our computerized schemes by comparing it with the subjective results of three experienced radiologists, and (5) assess the correlation, if any, between lung anatomy and lung function using statistical analyses. The unique advantages of this project include but are not limited to: (1) pulmonary fissure/lobe analysis is performed, analyzed and visualized in three-dimensional geometric space using computational geometry analysis and as a result have minimum dependence on anatomical knowledge of lung structure, and (2) lung functions may be studied non-invasively by analyzing pulmonary structures reconstructed from high resolution CT images on either a lobe-by-lobe or a whole lung basis. We are encouraged by the success of our preliminary studies that supports the feasibility of our approach.

Public Health Relevance

This project aims to ascertain information about the relationship, if any, between pulmonary fissures anatomy, including but not limited to fissure completeness and lung function through the development and testing of an objective, automated computerized detection and classification scheme. The long term goal is to enable an objective robust and consistent computerized approach to improve accuracy of early detection and classification (diagnosis) of lung disease and possibly to enable better management of patients as well as provide a useful tool for accurate early assessment of therapeutic efficacy.

Agency
National Institute of Health (NIH)
Institute
National Heart, Lung, and Blood Institute (NHLBI)
Type
Research Project (R01)
Project #
5R01HL096613-03
Application #
8257535
Study Section
Biomedical Imaging Technology Study Section (BMIT)
Program Officer
Gan, Weiniu
Project Start
2010-05-01
Project End
2014-04-30
Budget Start
2012-05-01
Budget End
2013-04-30
Support Year
3
Fiscal Year
2012
Total Cost
$224,978
Indirect Cost
$76,478
Name
University of Pittsburgh
Department
Radiation-Diagnostic/Oncology
Type
Schools of Medicine
DUNS #
004514360
City
Pittsburgh
State
PA
Country
United States
Zip Code
15213
Wang, Lei; Zhu, Jianbing; Sheng, Mao et al. (2018) Simultaneous segmentation and bias field estimation using local fitted images. Pattern Recognit 74:145-155
Wang, Xiaohua; Leader, Joseph K; Wang, Renwei et al. (2017) Vasculature surrounding a nodule: A novel lung cancer biomarker. Lung Cancer 114:38-43
Wang, Lei; Chang, Yan; Wang, Hui et al. (2017) An active contour model based on local fitted images for image segmentation. Inf Sci (N Y) 418-419:61-73
Wilson, David O; Pu, Jiantao (2016) The bell tolls for indeterminant lung nodules: computer-aided nodule assessment and risk yield (CANARY) has the wrong tune. J Thorac Dis 8:E836-7
Pu, Jiantao; Jin, Chenwang; Yu, Nan et al. (2015) A ""loop"" shape descriptor and its application to automated segmentation of airways from CT scans. Med Phys 42:3076-84
Gu, S; Li, R; Leader, J K et al. (2015) Obesity and extent of emphysema depicted at CT. Clin Radiol 70:e14-9
Ju, Jieyang; Li, Ruosha; Gu, Suicheng et al. (2014) Impact of emphysema heterogeneity on pulmonary function. PLoS One 9:e113320
Zhen, Yi; Gu, Suicheng; Meng, Xin et al. (2014) Automated identification of retinal vessels using a multiscale directional contrast quantification (MDCQ) strategy. Med Phys 41:092702
Gu, Suicheng; Meng, Xin; Sciurba, Frank C et al. (2014) Bidirectional elastic image registration using B-spline affine transformation. Comput Med Imaging Graph 38:306-14
Qiang, Yongqian; Wang, Qiuping; Xu, Guiping et al. (2014) Computerized segmentation of pulmonary nodules depicted in CT examinations using freehand sketches. Med Phys 41:041917

Showing the most recent 10 out of 25 publications