Keywords:
computed tomography, detectability index, image quality, imaging task, model-based iterative reconstruction, regularization design, task-driven imaging
This work investigates the prospective design of tube current modulation (TCM) and regularization for penalizedlikelihood(PL) reconstruction on an experimental cone-beam CT(CBCT) bench. The design follows a task driven imaging framework where a three-dimensional (3D) scout was first acquired and used as the anatomical model. A task-driven optimization process was then performed over a low-dimensional parameter space using the task-based image quality metric, detectability index (d'), as the objective function. Detectability index was computed as a function of a pre-specified imaging task and the noise and resolution properties of the reconstructed image predicted by a system model. The optimal combination of TCM and regularization was then used for subsequent data acquisition and reconstruction. The design was performed for a cluster discrimination task consisting of three 1/16” acetal spheres placed within a 3D printed elliptical phantom. Images were acquired on an experimental cone-beam CT system where TCM was achieved through a custom-modified interface for pulse width modulation at fixed tube current and tube voltage. In addition to imaging strategies dictated by the task-driven imaging framework, scans with no TCM (unmodulated) and conventional TCM based on variance minimization in FBP reconstruction were also performed at the same total exposure for comparison. The task driven imaging framework suggests that the optimal TCM for PL reconstruction has the opposite trend from that for FBP.Qualitative analysis of reconstructions for each strategy strongly support the trend of theoretical d' values. The TCM profile that is optimal for FBP performs poorly on low exposure data when used for PL. Task-driven strategies outperform the other methods providing the greatest number of visible stimuli in at low exposures. Initial investigations in this work show in experimental data that current, clinically implemented TCM strategies designed for FBP reconstruction can be suboptimal for model-based iterative reconstruction methods. Imaging performance is coupled to both the data acquisition and reconstruction methods,suggesting that current clinical protocols should be re-evaluated for newer model-based reconstruction approaches. The taskdriven imaging framework offers a promising approach for prospectively prescribing acquisition and reconstruction in a manner that maximizes task-based imaging performance for a given exposure.
- Grace J. Gang
- Johns Hopkins University
- Jeffrey H. Siewerdsen
- Johns Hopkins University
- Wenying Wang
- Johns Hopkins University
- J. Webster Stayman
- Johns Hopkins University
- Aswin Mathews
- Johns Hopkins University
- C. H. McCollough, A. N. Primak, N. Braun, J. Kofler, L.Yu, and J. Christner, “Strategies for Reducing Radiation Dose in CT,” Radiol. Clin. North Am., vol. 47, no. 1, pp. 27–40, 2009.
- P. Prakash, W. Zbijweski, G. J. Gang, Y. Ding, J. W. Stayman, J. Yorkston, J. A. A. Carrino, and J. H. Siewerdsen, “Task-based modeling and optimization of a cone-beam CT scanner for musculoskeletal imaging,” Med. Phys., vol. 38, no. 10, pp. 5612–5629, 2011.
- S. S. Hsieh and N. J. Pelc, “The feasibility of a piecewiselinear dynamic bowtie filter,” Med. Phys., vol. 40, no. 3, p.31910, 2013.
- T. P. Szczykutowicz and C. A. Mistretta, “Design of a digital beam attenuation system for computed tomography: part I. System design and simulation framework.,” Med. Phys., vol. 40, no. 2, p. 21905, Feb. 2013.
- J. W. Stayman, A. Mathews, W. Zbijweski, G. Gang, J. H. Siewerdsen, S. Kawamoto, I. Blevis, and R. Levinson, “Fluence-field modulated x-ray CT using multiple aperture devices,” in SPIE Medical Imaging, 2016, p. 97830X–97830X.
- M. K. M. Kalra, M. M. Maher, T. T. L. Toth, B. Schmidt, B. L. Westerman, H. T. Morgan, and S. Saini, “Techniques and applications of automatic tube current modulation for CT 1,” Radiology, vol. 233, no. 3, pp. 649–657, Dec. 2004.
- M. Mcnitt-gray, “Tube Current Modulation Approaches : Overview , Practical Issues and Potential Pitfalls Tube Current Modulation - Overview.” 2011.
- M. Beister, D. Kolditz, and W. A. Kalender, “Iterative reconstruction methods in X-ray CT,” Phys. medica, vol. 28, no. 2, pp. 94–108, 2012.
- P. J. Pickhardt, M. G. Lubner, D. H. Kim, J. Tang, J. A. Ruma, A. M. del Rio, and G.-H. Chen, “Abdominal CT with model-based iterative reconstruction (MBIR): initial results of a prospective trial comparing ultralow-dose with standard-dose imaging,” Am. J. Roentgenol., vol. 199, no. 6, pp. 1266–1274, 2012.
- M. Katsura, I. Matsuda, M. Akahane, J. Sato, H. Akai, K. Yasaka, A. Kunimatsu, and K. Ohtomo, “Model-based iterative reconstruction technique for radiation dose reduction in chest CT: comparison with the adaptive statistical iterative reconstruction technique,” Eur. Radiol., vol. 22, no. 8, pp. 1613–1623, 2012.
- J. W. Stayman and J. H. Siewerdsen, “Task-based trajectories in iteratively reconstructed interventional conebeam CT,” in Proceedings of 12th International Meeting on Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine, 2012, pp. 257–260.
- G. J. Gang, J. W. Stayman, T. Ehtiati, and J. H. Siewerdsen, “Task-driven image acquisition and reconstruction in conebeam CT.,” Phys. Med. Biol., vol. 60, no. 8, pp. 3129–50, Apr. 2015.
- G. J. Gang, J. W. Stayman, W. Zbijweski, and J. H. Siewerdsen, “Task-based detectability in CT image reconstruction by filtered backprojection and penalized likelihood estimation,” Med. Phys., vol. 41, no. 8, 2014.
- G. J. Gang, J. H. Siewerdsen, and J. W. Stayman, “Taskdriven tube current modulation and regularization design in computed tomography with penalized-likelihood reconstruction,” in SPIE Medical Imaging, 2016, p. 978324.
- M. Gies, W. A. Kalender, H. Wolf, C. Suess, and M. T. Madsen, “Dose reduction in CT by anatomically adapted tube current modulation. I. Simulation studies,” Med. Phys., vol. 26, no. 11, p. 2235, 1999.
- M. D. Harpen, “A simple theorem relating noise and patient dose in computed tomography,” Med. Phys., vol. 26, no. 11, p. 2231, 1999.
- G. J. Gang, J. H. Siewerdsen, and J. W. Stayman, “Taskdriven optimization of CT tube current modulation and regularization in model-based iterative reconstruction,” Phys. Med. Biol., vol. Submitted, 2016.
- J. A. Fessler, “Mean and variance of implicitly defined biased estimators (such as penalized maximum likelihood): applications to tomography.,” IEEE Trans. Image Process., vol. 5, no. 3, pp. 493–506, Jan. 1996.
- J. W. Stayman and J. A. Fessler, “Efficient calculation of resolution and covariance for penalized-likelihood reconstruction in fully 3-D SPECT.,” IEEE Trans. Med. Imaging, vol. 23, no. 12, pp. 1543–56, Dec. 2004.
- Y. Zhang-O’Connor and J. A. Fessler, “Fast predictions of variance images for fan-beam transmission tomography with quadratic regularization.,” IEEE Trans. Med. Imaging, vol. 26, no. 3, pp. 335–46, Mar. 2007.
- N. Hansen and S. Kern, “Evaluating the CMA evolution strategy on multimodal test functions,” in International Conference on Parallel Problem Solving from Nature, 2004, pp. 282–291.
- D. Kim, S. Ramani, and J. A. Fessler, “Accelerating X-ray CT ordered subsets image reconstruction with Nesterov’s first-order methods,” in Proc. Intl. Mtg. on Fully 3D Image Recon. in Rad. and Nuc. Med, 2013, pp. 22–25.
- G. J. Gang, J. H. Siewerdsen, and J. W. Stayman, “Task-Based Design of Fluence Field Modulation in CT for Model-Based Iterative Reconstruction,” in Proceedings of the 4th International Conference on Image Formation in X-Ray Computed Tomography, 2016, pp. 407–410.
- W. Wang, G. J. Gang, and J. W. Stayman, “Spatial mresolution properties in penalized-likelihood reconstruction of blurred tomographic data,” in Fully 3D Meeting, 2017, p. Submitted.