Esophageal wall thickness is an important predictor of esophageal cancer response to therapy. In this study, we developed a computerized pipeline for quantification of esophageal wall thickness using computerized tomography (CT). We first segmented the esophagus using a multi-atlas-based segmentation scheme. The esophagus in each atlas CT was manually segmented to create a label map. Using image registration, all of the atlases were aligned to the imaging space of the target CT. The deformation field from the registration was applied to the label maps to warp them to the target space. A weighted majority-voting label fusion was employed to create the segmentation of esophagus. Finally, we excluded the lumen from the esophagus using a threshold of -600 HU and measured the esophageal wall thickness. The developed method was tested on a dataset of 30 CT scans, including 15 esophageal cancer patients and 15 normal controls. The mean Dice similarity coefficient (DSC) and mean absolute distance (MAD) between the segmented esophagus and the reference standard were employed to evaluate the segmentation results. Our method achieved a mean Dice coefficient of 65.55 ± 10.48% and mean MAD of 1.40 ± 1.31 mm for all the cases. The mean esophageal wall thickness of cancer patients and normal controls was 6.35 ± 1.19 mm and 6.03 ± 0.51 mm, respectively. We conclude that the proposed method can perform quantitative analysis of esophageal wall thickness and would be useful for tumor detection and tumor response evaluation of esophageal cancer.
Golden retriever muscular dystrophy (GRMD) is a canine model of Duchenne muscular dystrophy (DMD) that has been
increasingly used in both pathogenetic and therapeutic pre-clinical studies. Recent studies have shown that Magnetic
resonance imaging (MRI) has great potential to noninvasively assess muscle disorders and has been increasingly used to monitor disease progression in DMD patients and GRMD dogs. In this study, we developed a statistical texture analysis based MRI quantification framework for GRMD. Our system was applied to a database of 45 MRI scans from 8 normal and 10 GRMD dogs in a natural history study. The dogs were longitudinally scanned at 3, 6 and 9 months of age. We first segmented six proximal limb muscles of each dog using a semi-automated, interpolation-based method and then automatically measured the 3D first-order histogram and novel 3D high-order run-length matrix based texture features within each segmented muscle. Our results indicated that MRI texture features has the ability to distinguish the normal and GRMD muscles at each age. Our experimental results demonstrated the potential of MRI texture measurements to serve as biomarkers to distinguish normal and muscular dystrophic muscles in DMD patients.
We are developing an automated method for detection and quantification of ischemic stroke in computed tomography (CT). Ischemic stroke often connects to brain ventricle, therefore, ventricular segmentation is an important and difficult task when stroke is present, and is the topic of this study. We first corrected inclination angle of brain by aligning midline of brain with the vertical centerline of a slice. We then estimated the intensity range of the ventricles by use of the k-means method. Two segmentation of the ventricle were obtained by use of thresholding technique. One segmentation contains ventricle and nearby stroke. The other mainly contains ventricle. Therefore, the stroke regions can be extracted and removed using image difference technique. An adaptive template-matching algorithm was employed to identify objects in the fore-mentioned segmentation. The largest connected component was identified and considered as the ventricle. We applied our method to 25 unenhanced CT scans with stroke. Our method achieved average Dice index, sensitivity, and specificity of 95.1%, 97.0%, and 99.8% for the entire ventricular regions. The experimental results demonstrated that the proposed method has great potential in detection and quantification of stroke and other neurologic diseases.
Duchenne muscular dystrophy (DMD) is a progressive and fatal X-linked disease caused by mutations in the DMD gene.
Magnetic resonance imaging (MRI) has shown potential to provide non-invasive and objective biomarkers for
monitoring disease progression and therapeutic effect in DMD. In this paper, we propose a semi-automated scheme to
quantify MRI features of golden retriever muscular dystrophy (GRMD), a canine model of DMD. Our method was
applied to a natural history data set and a hydrodynamic limb perfusion data set. The scheme is composed of three
modules: pre-processing, muscle segmentation, and feature analysis. The pre-processing module includes: calculation of
T2 maps, spatial registration of T2 weighted (T2WI) images, T2 weighted fat suppressed (T2FS) images, and T2 maps,
and intensity calibration of T2WI and T2FS images. We then manually segment six pelvic limb muscles. For each of the
segmented muscles, we finally automatically measure volume and intensity statistics of the T2FS images and T2 maps.
For the natural history study, our results showed that four of six muscles in affected dogs had smaller volumes and all
had higher mean intensities in T2 maps as compared to normal dogs. For the perfusion study, the muscle volumes and
mean intensities in T2FS were increased in the post-perfusion MRI scans as compared to pre-perfusion MRI scans, as
predicted. We conclude that our scheme successfully performs quantitative analysis of muscle MRI features of GRMD.
We developed an automated method for the segmentation of lungs with severe diffuse interstitial lung disease (DILD) in
multi-detector CT. In this study, we would like to compare the performance levels of this method and a thresholdingbased
segmentation method for normal lungs, moderately abnormal lungs, severely abnormal lungs, and all lungs in our
database. Our database includes 31 normal cases and 45 abnormal cases with severe DILD. The outlines of lungs were
manually delineated by a medical physicist and confirmed by an experienced chest radiologist. These outlines were used
as reference standards for the evaluation of the segmentation results. We first employed a thresholding technique for CT
value to obtain initial lungs, which contain normal and mildly abnormal lung parenchyma. We then used texture-feature
images derived from co-occurrence matrix to further segment lung regions with severe DILD. The segmented lung
regions with severe DILD were combined with the initial lungs to generate the final segmentation results. We also
identified and removed the airways to improve the accuracy of the segmentation results. We used three metrics, i.e.,
overlap, volume agreement, and mean absolute distance (MAD) between automatically segmented lung and reference
lung to evaluate the performance of our segmentation method and the thresholding-based segmentation method. Our
segmentation method achieved a mean overlap of 96.1%, a mean volume agreement of 98.1%, and a mean MAD of 0.96
mm for the 45 abnormal cases. On the other hand the thresholding-based segmentation method achieved a mean overlap
of 94.2%, a mean volume agreement of 95.8%, and a mean MAD of 1.51 mm for the 45 abnormal cases. Our new
method obtained higher performance level than the thresholding-based segmentation method.
The successful development of high performance computer-aided-diagnostic systems has potential to assist radiologists
in the detection and diagnosis of diffuse lung disease. We developed in this study an automated scheme for the detection
of diffuse lung disease on multi-detector computed tomography (MDCT). Our database consisted of 68 CT scans, which
included 31 normal and 37 abnormal cases with three kinds of abnormal patterns, i.e., ground glass opacity, reticular,
and honeycombing. Two radiologists first selected the CT scans with abnormal patterns based on clinical reports. The
areas that included specific abnormal patterns in the selected CT images were then delineated as reference standards by
an expert chest radiologist. To detect abnormal cases with diffuse lung disease, the lungs were first segmented from the
background in each slice by use of a texture analysis technique, and then divided into contiguous volumes of interest
(VOIs) with a 64×64×64 matrix size. For each VOI, we calculated many statistical texture features, including the mean
and standard deviation of CT values, features determined from the run length matrix, and features from the co-occurrence
matrix. A quadratic classifier was employed for distinguishing between normal and abnormal VOIs by use of
a leave-one-case-out validation scheme. A rule-based criterion was employed to further determine whether a case was
normal or abnormal. For the detection of abnormal VOIs, our CAD system achieved a sensitivity of 86% and a
specificity of 90%. For the detection of abnormal cases, it achieved a sensitivity of 89% and a specificity of 90%. This
preliminary study indicates that our CAD system would be useful for the detection of diffuse lung disease.
KEYWORDS: Tumors, Image segmentation, Brain, 3D image processing, Magnetic resonance imaging, Neuroimaging, Solids, 3D image reconstruction, Computer programming, 3D magnetic resonance imaging
Accurate volumetry of brain tumors in magnetic resonance imaging (MRI) is important for evaluating the interval changes in tumor volumes during and after treatment, and also for planning of radiation therapy. In this study, an automated volumetry method for brain tumors in MRI was developed by use of a new three-dimensional (3-D) image segmentation technique. First, the central location of a tumor was identified by a radiologist, and then a volume of interest (VOI) was determined automatically. To substantially simplify tumor segmentation, we transformed the 3-D image of the tumor into a two-dimensional (2-D) image by use of a "spiral-scanning" technique, in which a radial line originating from the center of the tumor scanned the 3-D image spirally from the "north pole" to the "south pole". The voxels scanned by the radial line provided a transformed 2-D image. We employed dynamic programming to delineate an "optimal" outline of the tumor in the transformed 2-D image. We then transformed the optimal outline back into 3-D image space to determine the volume of the tumor. The volumetry method was trained and evaluated by use of 16 cases with 35 brain tumors. The agreement between tumor volumes provided by computer and a radiologist was employed as a performance metric. Our method provided relatively accurate results with a mean agreement value of 88&percent;.
Lung nodule segmentation in computed tomography (CT) plays an important role in computer-aided detection, diagnosis,
and quantification systems for lung cancer. In this study, we developed a simple but accurate nodule segmentation
method in three-dimensional (3D) CT. First, a volume of interest (VOI) was determined at the location of a nodule. We
then transformed the VOI into a two-dimensional (2D) image by use of a "spiral-scanning" technique, in which a radial
line originating from the center of the VOI spirally scanned the VOI. The voxels scanned by the radial line were
arranged sequentially to form a transformed 2D image. Because the surface of a nodule in 3D image became a curve in
the transformed 2D image, the spiral-scanning technique considerably simplified our segmentation method and enabled
us to obtain accurate segmentation results. We employed a dynamic programming technique to delineate the "optimal"
outline of a nodule in the 2D image, which was transformed back into the 3D image space to provide the interior of the
nodule. The proposed segmentation method was trained on the first and was tested on the second Lung Image Database
Consortium (LIDC) datasets. An overlap between nodule regions provided by computer and by the radiologists was
employed as a performance metric. The experimental results on the LIDC database demonstrated that our segmentation
method provided relatively robust and accurate segmentation results with mean overlap values of 66% and 64% for the
nodules in the first and second LIDC datasets, respectively, and would be useful for the quantification, detection, and
diagnosis of lung cancer.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.