Computerized Medical Imaging and Graphics
Evaluation of the quantitative capability of a home-made cone-beam micro computed tomography system
Introduction
Among osteological diseases, osteoporosis is common in climacteric woman [1] and osteoarthritis is a common age-related joint disease [2]. An important issue of osteopathic studies is bone mineral density (BMD), which measures the amount of calcium in various regions of the bones. Many modalities can carry out a BMD test and these include quantitative ultrasound [3], dual energy X-ray absorptiometry (DEXA) [4], [5], single energy X-ray absorptiometry (SXA), peripheral dual energy X-ray absorptiometry (PDXA), radiographic absorptiometry (RA), dual photon absorptiometry (DPA), single photon absorptiometry (SPA) and quantitative computed tomography (QCT) [4], [6], [7], [8], [9], [10]. Among these modalities, QCT is the most accurate. However, QCT is not widely used because it delivers a higher radiation dose to patients than ultrasound or DEXA. Even so, QCT has significant potential in animal studies where the dose is of less concerned.
Among small-sample imaging modalities, micro computed tomography, often called microCT, is utilized to obtain high-resolution tomographic images. The microCT system used in this study is a home-made cone-beam computed tomography apparatus, which provides volume information from multiple two-dimensional projections as shown in Fig. 1. Compared to a traditional computed tomography system, the cone-beam CT has a major advantage in three-dimensional imaging because the cone-beam CT can produce volume information directly without stacking multiple two-dimensional cross sections. Therefore, cone-beam CT is able to rapidly acquire high-resolution three-dimensional images and is superior for small animal imaging.
The geometry employed in this study for the microCT is a circular orbit. Unfortunately, an exact reconstruction condition is difficult to obtain in this arrangement. However, several approximate reconstruction methods have been proposed. The most popularly one is the “practical cone-beam algorithm”, which was developed by Feldkamp et al. [11] and is also called the FDK algorithm or the Feldkamp algorithm. Reconstruction algorithms can be roughly categorized into two classes: analytic and iterative. The Feldkamp algorithm is an analytic reconstruction. It is a successful and efficient reconstruction algorithm. Nevertheless, the Feldkamp algorithm is limited by circular scanning, spherical specimen reconstruction and longitudinal image blurring. For these reasons, modified Feldkamp approaches are used in practice, such as the general cone-beam reconstruction algorithm [12] and the T-FDK algorithm [13]. Compared to the analytic method, iterative reconstruction, such as the maximum likelihood algorithm (ML) [14], [15] and expectation maximization algorithm [16], have the potential to obtain a higher quality image, but demand more computing resources than the analytic algorithms. In this study, results obtained by T-FDK [17] and ordered subsets maximum likelihood (ML) algorithms were obtained and evaluated.
Section snippets
The T-FDK algorithm
The set of the cone-beam projection data R(α, c, r) for detector measurement along a circular source–detector trajectory is parameterized by the source angle on the circular trajectory α ∈ [0, 2π], the column coordinate on the detector c ∈ [−cmax, +cmax] and the row coordinate on the detector r ∈ [−rmax, +rmax]. An illustration of the acquisition geometry of the original cone-beam projection data is shown in Fig. 2.
In the T-FDK algorithm, the original cone-beam projection data can be rebinned to form
Materials and methods
In our microCT system, the object is placed on a rotating stage and is illuminated by a source–detector pair over 360° to acquire 400 cone-beam projections. A schematic view is presented in Fig. 1. A GOS-CMOS detector is utilized to detect the X-ray signals. The detector size is 5 cm × 5 cm and the matrix size is 1024 × 1024 pixels. The field of view is 4 cm diameter. The isotropic pixel is 39 μm.
As is well known, the intensity of the CT image depends on the mass attenuation coefficient and is
Results
The HA phantom dataset with and without correction was reconstructed using the T-FDK and ML-OS algorithms. The four different images reconstructions from the same data are shown in Fig. 6. Every image is shown using the same window level and width. The contrast of the images after beam hardening correction is obviously higher than when the images are uncorrected. Four intensity curves against density are shown in Fig. 7. The intensities have a mean value for the five regions of interest and
Discussion and conclusions
Most commercial CT systems use analytic reconstruction. When either the analytic or iterative reconstruction is applied, it is based on the system being monochromatic and the reconstructed image will show a worse accuracy if the projections are polychromatic. A polychromatic analytic reconstruction does not exist so far. Traditional CT uses a beam hardening correction to obtain simulated monochromatic projections and then the dataset is then reconstructed by analytic reconstruction. The problem
Acknowledgments
The research was supported by grants (NSC 93-2622-B-010-002-CC3 and NSC 94-2622-B-010-003-CC3) from National Science Council, Taiwan. We thank Brian Shieh for helpful discussion.
Ho-Shiang Chueh received the MS degree in Radiological Sciences from National Yang-Ming University, Taiwan, R.O.C. in 2003. He is currently a PhD student at National Yang-Ming University. His research interests include statistical image reconstruction and animal CT development.
References (24)
- et al.
Development and pharmacological characterization of a rat model of osteoarthritis pain
Pain
(2005) - et al.
Calcitonin therapy in osteoporosis
Treat Endocrinol
(2004) - et al.
Calcaneal ultrasound predicts early postmenopausal fractures as well as axial BMD. A prospective study of 422 women
Osteoporos Int
(2004) - et al.
Comparison of DXA, QCT and trabecular structure in beta-thalassaemia
Eur J Haematol
(2005) - et al.
Comparison of bone mass in forearm, lumbar vertebra and hip by single and/or dual energy X-ray absorptiometry
Chin Med Sci J
(1999) - et al.
Normal bone mineral content levels determined by quantitative computer tomography (QCT) in the population of Moldavia
Rev Med Chir Soc Med Nat Iasi
(2004) - et al.
Heritability of spinal trabecular volumetric bone mineral density measured by QCT in the Diabetes Heart Study
Calcif Tissue Int
(2004) - et al.
Differences in hip quantitative computed tomography (QCT) measurements of bone mineral density and bone strength between glucocorticoid-treated and glucocorticoid-naive postmenopausal women
Osteoporos Int
(2005) - et al.
DXA vs QCT for subclinical celiac disease patients
Acta Diabetol
(2003) - et al.
DXA vs QCT: in vitro and in vivo studies
Acta Diabetol
(2003)
Practical cone-beam algorithm
J Opt Soc Am
A general cone-beam reconstruction algorithm
IEEE Trans Med Imaging
Cited by (7)
Specimen size and porosity can introduce error into μCT-based tissue mineral density measurements
2009, BoneCitation Excerpt :Recently, methods for the measurement of ρm have been developed for polychromatic μCT and several studies have incorporated these techniques [9–22]. However, only a few published studies of the precision and accuracy of μCT-based equivalent ρm have been performed to date [23–26], in contrast to the numerous studies completed for clinical CT [1–6,27–38]. The accuracy and precision of μCT-based measurements of ρm can be affected by factors related to the scan settings, tissue samples, and scan artifacts.
Medical imaging and telemedicine - from medical data production, to processing, storing, and sharing: A short outlook
2006, Computerized Medical Imaging and GraphicsUremia Induces Dental Pulp Ossification but Reciprocally Inhibits Adjacent Alveolar Bone Osteogenesis
2015, Calcified Tissue InternationalApplication of polychromatic μcT for mineral density determination
2011, Journal of Dental ResearchA system of calibrating microtomography for use in caries research
2009, Caries Research
Ho-Shiang Chueh received the MS degree in Radiological Sciences from National Yang-Ming University, Taiwan, R.O.C. in 2003. He is currently a PhD student at National Yang-Ming University. His research interests include statistical image reconstruction and animal CT development.
Jyh-Cheng Chen received the BS degree in physics from National Central University, Taiwan, R.O.C., in 1983, and the MS degree in physics and PhD degree in optical sciences from the University of Arizona in 1988 and 1995, respectively. In 1995, he joined the Opto-Electronics and System Laboratories, Industrial Technology Research Institute, Taiwan, R.O.C. as a research associate working on semiconductor laser packaging. In 1996, he became a member of the faculty of Division of Radiological Science and Technology, Department of Medical Technology, National Yang-Ming University, Taiwan, R.O.C. In 1998, he became an associate professor of Institute of Radiological Sciences, National Yang-Ming University, Taiwan. Since 2005, he has been a professor in the Department of Biomedical Imaging and Radiological Sciences, National Yang-Ming University, Taiwan, teaching and pursuing his research interests in areas of radiological imaging and nuclear medicine instrumentation. In particular, Prof. Chen is using microPET to do statistical image reconstruction, processing and analysis for animal molecular imaging studies. Currently he is promoting the NYMU Small-animal Gamma-ray Imaging Laboratory (animal SPECT/CT). Prof. Chen is a member of Society of Nuclear Medicine, Society for Molecular Imaging, and the Institute of Electrical and Electronics Engineering (IEEE).