ETD

Archivio digitale delle tesi discusse presso l'Università di Pisa

Tesi etd-09262017-234830


Tipo di tesi
Tesi di laurea magistrale
Autore
DI TRAPANI, VITTORIO
URN
etd-09262017-234830
Titolo
Characterization of a system for breast-CT with synchrotron radiation
Dipartimento
FISICA
Corso di studi
FISICA
Relatori
relatore Prof. Delogu, Pasquale
Parole chiave
  • computed tomography (CT)
  • breast-CT
  • CdTe detector
Data inizio appello
18/10/2017
Consultabilità
Completa
Riassunto
Early detection of breast cancer greatly increases the chances for successful treatments reducing mortality by at least 20% [1]. To date, although mammography is the main tool for detecting breast cancer, it poses several limitations to the detectability of tumors due to the superimposition of breast glandular structures [2].

Various studies have shown that breast-computed tomography (CT) increases the sen-sitivity of breast cancer detectability eliminating superimposition [3]. Moreover, it has been demonstrated that breast-CT improves contrast resolution compared to mammog-raphy, leading to a better identification and classification of breast tissues at the cost of an increased dose [4] [5].

In the past, the main disadvantages of breast-CT (using whole-body CT scanners) were low spatial resolution and high dose to the patient compared to mammography [6]. Recently, the increased interest in breast-CT among radiologists led to the implementation of dedicated breast-CT scanners able to meet the needs for high spatial and contrast res-olution. In particular, a dedicated breast-CT imaging system with synchrotron radiation at the SYRMEP (Synchrotron Radiation for Medical Physics) beamline of Elettra (Tri-este) was developed through the SYRMA-3D (SYnchrotron Radiation Mammography-3D) project.

SYRMA-3D uses ‘Pixirad-8’, a direct detection photon counting device with CdTe sensor and pixel size of 60 m, which potentially provides improved performances in terms of noise, efficiency and spatial resolution.

The quality of CT images depends mainly on the source, the detector and the recon-struction software.

This work focuses on the characterization of the detection system and on the opti-mization of the CT reconstructions using the objective metrics for the evaluation of the spatial resolution and the noise.

The detector imaging properties have been assessed from planar images. In frequency domain (FD), the noise has been evaluated by means of the Normalized Noise Power Spec-trum (NNPS). The spatial resolution has been measured with an innovative technique, which allows analysis at single pixel level. The presampling Edge Spread Function (pESF) and the presampling Line Spread Function (pLSF) have been employed as metrics in the spatial domain (SD), while the presampling Modulation Transfer Function (pMTF) has been used in FD. This analysis has shown how the imaging properties of this system can be influenced by physical limitations1, specific pixel geometry and resampling process.

The choice of the algorithm and of the parameters of reconstruction for CT-images, such as the voxel size, are not univocal. For this system, the optimal algorithm for CT reconstructions has been investigated between the Filtered Back Projection (FBP) with different filters, the Simultaneous Algebraic Reconstruction Technique (SART) and the Simultaneous Iterative Reconstruction Technique (SIRT). CT images have been recon-structed, using two voxel sizes (V60 = 603 m3 and V120 = 1203 m3), with ASTRA toolbox (All Scales Tomographic Reconstruction Antwerp), an open source software char-acterized by its modularity and customization [7].

For CT reconstructions, noise and spatial resolution have been assessed through the NNPS (measured on a homogeneous phantom [8]) and the Point Spread Function (PSF), the LSF and the MTF (measured with the thin wire method [9], [10]).

Finally, the measurements of the NNPS and the MTF have been used to calculate the Noise Equivalent number of Quanta (NEQ). This quantity summarizes the informations about the spatial resolution and the noise and can be used to directly compare the overall quality of the images. In particular, NEQ has been used to directly compare the CT reconstructions performed in different conditions. This analysis allowed to characterize the performances of the reconstruction algorithms. This provides a guide for choosing the best algorithm and voxel size to be used, in order to achieve optimal performances in terms of spatial resolution and noise.

References

[1] World Health Organization regional office for Europe website:

http://www.euro.who.int

[2] O’Connell, Avice M. and Karellas, Andrew and Vedantham, Srinivasan, The Potential Role of Dedicated 3D Breast CT as a Diagnostic Tool: Review and Early Clinical Examples, The Breast Journal, vol.20, nr.6, 2014.

[3] Boone JM, Kwan AL, Yang K, et al. Computed tomography for imaging the breast. J Mammary Gland Biol Neoplasia. 2006;11:103-111.

[4] Boone JM, Nelson TR, Lindfors KK, et al. 2001, Dedicated breast CT: radiation dose and image quality evaluation. Radiology 221:657-667.

[5] Nelson TR, Cervina LI, Boone JM. Classification of breast computed tomography data. Med Phys. 2008;35:1078-1086.

[6] Glick S. Breast CT. Annu Rev Biomed Eng. 2007;9:501-526.

[7] W. Van Aarle, W.J.Palenstijin, J. De Beenhouwer, T. Atlantzis, S. Bals, K. J. Baten-burg and J. Sijbers, Fast and Flexible X-ray Tomography Using the ASTRA toolbox, Optics Express, 24(22), 25129-25147, 2016.

[8] S. N. Friedman, G. S. K. Fung, J. H. Siewerdsen, and B. M. W. Tsui. "A simple approach to measure computed tomography (CT) modulation transfer function (MTF) and noise-power spectrum (NPS) using the American College of Radiology (ACR) accreditation phantom" Med. Phys. 40, 051907-1 - 051907-9 (2013).

[9] JT Bushberg, JA Seibert, EM Leidholdt Jr. and JM Boone, The Essential Physics of Medical Imaging, 2nd ed., Philadelphia: Lippincott Williams & Wilkins, 2002.

[10] X Tang, S Narayanan, J Hsieh, JD Pack, SM Mcolash, P Sainath, RA Nilsen and B Taha, Enhancement of in-plane spatial resolution in volumetric Computed Tomog-raphy with focal spot wobbling – Overcoming the constraint on number of projection views per gantry rotation, J. X-ray Sci. Technol., v.18, pp. 251-65, 2010.
File