Abstract
Computed tomography (CT), a medical imaging technique, offers a detailed view of
the human body that can be used for direct inspection of the lung tissue. This allows
for in vivo measurement of subtle disease patterns such as the patterns associated
with chronic obstructive pulmonary disease (COPD). COPD, also commonly referred
to as “smokers’ lungs”, is a lung disease characterized by limitation of the airflow to
and from the lungs causing shortness of breath. The disease is expected to rank as the
fifth most burdening disease worldwide by 2020 according the the World Health Organization.
COPD comprises two main components, chronic bronchitis, characterized
by inflammation in the airways, and emphysema, characterized by loss of lung tissue.
Emphysema basically looks like black blobs of varying sizes within the normal, gray
lung tissue in CT, and can therefore be seen as a family of texture patterns. Commonly
employed CT-based quantitative measures in the clinical literature are rather
simplistic and do not take the texture appearance of the lung tissue into account. This
includes measures such as the relative area (RA), also called emphysema index, that
applies a fixed threshold to each individual lung voxel in the CT image and counts
the number of voxels below the threshold relative to the total amount of lung voxels.
This thesis presents several methods for texture-based quantification of emphysema
and/or COPD in CT images of the lungs. The methods rely on image processing
and pattern recognition. The image processing part deals with characterizing the lung
tissue texture using a suitable texture descriptor. Two types of descriptors are considered,
the local binary pattern histogram and histograms of filter responses from a
multi-scale Gaussian derivative filter bank. The pattern recognition part is used to
turn the texture measures, measured in a CT image of the lungs, into a quantitative
measure of disease. This is done by applying a classifier that is trained on a training
set of data examples with known lung tissue patterns. Different classification systems
are considered, and we will in particular use the pattern recognition concepts of
supervised learning, multiple instance learning, and dissimilarity representation-based
classification.
The proposed texture-based measures are applied to CT data from two different
sources, one comprising low dose CT slices from subjects with manually annotated
regions of emphysema and healthy tissue, and one comprising volumetric low dose CT
images from subjects that are either healthy or suffer from COPD. Several experiments
demonstrate that it is clearly beneficial to take the lung tissue texture into account
when classifying or quantifying emphysema and/or COPD in CT. Compared to RA
and other common clinical CT-based measures, the texture-based measures are better
at discriminating between CT images from healthy and COPD subjects, they correlate
better with the lung function of the subjects, they are more reproducible, and they
are less influence by the inspiration level of the subject during CT scanning – a major
source of variability in CT.
the human body that can be used for direct inspection of the lung tissue. This allows
for in vivo measurement of subtle disease patterns such as the patterns associated
with chronic obstructive pulmonary disease (COPD). COPD, also commonly referred
to as “smokers’ lungs”, is a lung disease characterized by limitation of the airflow to
and from the lungs causing shortness of breath. The disease is expected to rank as the
fifth most burdening disease worldwide by 2020 according the the World Health Organization.
COPD comprises two main components, chronic bronchitis, characterized
by inflammation in the airways, and emphysema, characterized by loss of lung tissue.
Emphysema basically looks like black blobs of varying sizes within the normal, gray
lung tissue in CT, and can therefore be seen as a family of texture patterns. Commonly
employed CT-based quantitative measures in the clinical literature are rather
simplistic and do not take the texture appearance of the lung tissue into account. This
includes measures such as the relative area (RA), also called emphysema index, that
applies a fixed threshold to each individual lung voxel in the CT image and counts
the number of voxels below the threshold relative to the total amount of lung voxels.
This thesis presents several methods for texture-based quantification of emphysema
and/or COPD in CT images of the lungs. The methods rely on image processing
and pattern recognition. The image processing part deals with characterizing the lung
tissue texture using a suitable texture descriptor. Two types of descriptors are considered,
the local binary pattern histogram and histograms of filter responses from a
multi-scale Gaussian derivative filter bank. The pattern recognition part is used to
turn the texture measures, measured in a CT image of the lungs, into a quantitative
measure of disease. This is done by applying a classifier that is trained on a training
set of data examples with known lung tissue patterns. Different classification systems
are considered, and we will in particular use the pattern recognition concepts of
supervised learning, multiple instance learning, and dissimilarity representation-based
classification.
The proposed texture-based measures are applied to CT data from two different
sources, one comprising low dose CT slices from subjects with manually annotated
regions of emphysema and healthy tissue, and one comprising volumetric low dose CT
images from subjects that are either healthy or suffer from COPD. Several experiments
demonstrate that it is clearly beneficial to take the lung tissue texture into account
when classifying or quantifying emphysema and/or COPD in CT. Compared to RA
and other common clinical CT-based measures, the texture-based measures are better
at discriminating between CT images from healthy and COPD subjects, they correlate
better with the lung function of the subjects, they are more reproducible, and they
are less influence by the inspiration level of the subject during CT scanning – a major
source of variability in CT.
| Original language | English |
|---|
| Place of Publication | København |
|---|---|
| Publisher | Faculty of Science, University of Copenhagen |
| Publication status | Published - 2010 |