Efficient Hyper-Parameter Selection in Total Variation-Penalised XCT Reconstruction Using Freund and Shapire’s Hedge Approach
Abstract
1. Introduction
1.1. Inverse Problems, Regularisation and Hyperparameter Tuning
1.2. Algorithms for XCT with Few Projections
1.3. The Problem of Hyperparameter Calibration
2. The Adaptive-Weighted Projection-Controlled Steepest Descent of Lohvithee et al.
2.1. Description of the Reconstruction Method
Algorithm 1: Adaptive-weighted Projection-Controlled Steepest Descent (AwPCSD) |
2.2. Hyper-Parameters and Stopping Criterion of the AwPCSD Algorithm
2.2.1. Data-inconsistency-tolerance Parameter
2.2.2. TV sub-iteration Number ()
2.2.3. Relaxation Parameter ()
2.2.4. Reduction Factor of Relaxation Parameter ()
2.2.5. Scale Factor for Adaptive-weighted TV Norm ()
3. Hedging Parameter Selection
3.1. The Proposed Approach
- A prediction is performed for the next observed projection and an error is measured for every ,
- A loss is computed for each value ,
- A probability, denoted by , is associated with each and the probability vector is updated according to the rule [27]:
Algorithm 2: Hedge based selection |
3.2. The Cross-Validation Approach
4. Numerical Studies
4.1. Experiments with Digital 4D Extended Cardiac-Torso (XCAT) Phantom
4.2. Performance Evaluation
4.3. Experiments with Different Datasets
4.4. Experiments with Real Data
5. Conclusions
Author Contributions
Funding
Conflicts of Interest
References
- Candès, E.J.; Romberg, J.; Tao, T. Robust uncertainty principles: Exact signal reconstruction from highly incomplete frequency information. IEEE Trans. Inf. Theory 2006, 52, 489–509. [Google Scholar] [CrossRef]
- Donoho, D.L. Compressed sensing. IEEE Trans. Inf. Theory 2006, 52, 1289–1306. [Google Scholar] [CrossRef]
- Candès, E.J. Compressive sampling. In Proceedings of the International Congress of Mathematicians, Madrid, Spain, 22–30 August 2006; Volume 3, pp. 1433–1452. [Google Scholar]
- Candès, E.J.; Wakin, M.B. An introduction to compressive sampling. IEEE Signal Process. Mag. 2008, 25, 21–30. [Google Scholar] [CrossRef]
- Chrétien, S.; Darses, S. Sparse recovery with unknown variance: a LASSO-type approach. IEEE Trans. Inf. Theory 2014, 60, 3970–3988. [Google Scholar] [CrossRef]
- Bredies, K.; Kunisch, K.; Pock, T. Total generalized variation. SIAM J. Imaging Sci. 2010, 3, 492–526. [Google Scholar] [CrossRef]
- Goldstein, T.; Osher, S. The split Bregman method for L1-regularized problems. SIAM J. Imaging Sci. 2009, 2, 323–343. [Google Scholar] [CrossRef]
- Chambolle, A. An algorithm for total variation minimization and applications. J. Math. Imaging vis. 2004, 20, 89–97. [Google Scholar]
- Feldkamp, L.; Davis, L.; Kress, J. Practical cone-beam algorithm. JOSA A 1984, 1, 612–619. [Google Scholar] [CrossRef]
- Rudin, L.I.; Osher, S.; Fatemi, E. Nonlinear total variation based noise removal algorithms. Phys. D Nonlinear Phenom. 1992, 60, 259–268. [Google Scholar] [CrossRef]
- Sidky, E.Y.; Kao, C.M.; Pan, X. Accurate image reconstruction from few-views and limited-angle data in divergent-beam CT. J. X-ray Sci. Technol. 2006, 14, 119–139. [Google Scholar]
- Sidky, E.Y.; Pan, X. Image reconstruction in circular cone-beam computed tomography by constrained, total-variation minimization. Phys. Med. Biol. 2008, 53, 4777. [Google Scholar] [CrossRef] [PubMed]
- Liu, L.; Yin, Z.; Ma, X. Nonparametric optimization of constrained total variation for tomography reconstruction. Comput. Biol. Med. 2013, 43, 2163–2176. [Google Scholar] [CrossRef] [PubMed]
- Lu, X.; Sun, Y.; Yuan, Y. Optimization for limited angle tomography in medical image processing. Pattern Recognit. 2011, 44, 2427–2435. [Google Scholar] [CrossRef]
- Tian, Z.; Jia, X.; Yuan, K.; Pan, T.; Jiang, S.B. Low-dose CT reconstruction via edge-preserving total variation regularization. Phys. Med. Biol. 2011, 56, 5949. [Google Scholar] [CrossRef]
- Liu, L.; Lin, W.; Jin, M. Reconstruction of sparse-view X-ray computed tomography using adaptive iterative algorithms. Comput. Biol. Med. 2015, 56, 97–106. [Google Scholar] [CrossRef]
- Grudzien, K.; Romanowski, A.; Williams, R.A. Application of a Bayesian approach to the tomographic analysis of hopper flow. Part. Part. Syst. Charact. 2005, 22, 246–253. [Google Scholar] [CrossRef]
- Liu, Y.; Ma, J.; Fan, Y.; Liang, Z. Adaptive-weighted total variation minimization for sparse data toward low-dose x-ray computed tomography image reconstruction. Phys. Med. Biol. 2012, 57, 7923. [Google Scholar] [CrossRef]
- Perona, P.; Malik, J. Scale-space and edge detection using anisotropic diffusion. IEEE Trans. Pattern Anal. Mach. Intell. 1990, 12, 629–639. [Google Scholar] [CrossRef]
- Wang, J.; Li, T.; Liang, Z.; Xing, L. Dose reduction for kilovotage cone-beam computed tomography in radiation therapy. Phys. Med. Biol. 2008, 53, 2897. [Google Scholar] [CrossRef]
- Lohvithee, M.; Biguri, A.; Soleimani, M. Parameter selection in limited data cone-beam CT reconstruction using edge-preserving total variation algorithms. Phys. Med. Biol. 2017, 62, 9295. [Google Scholar] [CrossRef]
- Claesen, M.; De Moor, B. Hyperparameter search in machine learning. arXiv preprint 2015, arXiv:1502.02127. [Google Scholar]
- Padierna, L.C.; Carpio, M.; Rojas, A.; Puga, H.; Baltazar, R.; Fraire, H. Hyper-parameter tuning for support vector machines by estimation of distribution algorithms. In Nature-Inspired Design of Hybrid Intelligent Systems; Springer: New York, NY, USA, 2017; pp. 787–800. [Google Scholar]
- Lorenzo, P.R.; Nalepa, J.; Kawulok, M.; Ramos, L.S.; Pastor, J.R. Particle swarm optimization for hyper-parameter selection in deep neural networks. In Proceedings of the Genetic and Evolutionary Computation Conference, Berlin, Germany, 15–19 July 2017; pp. 481–488. [Google Scholar]
- Palar, P.S.; Zuhal, L.R.; Shimoyama, K. On the use of metaheuristics in hyperparameters optimization of gaussian processes. In Proceedings of the Genetic and Evolutionary Computation Conference Companion, Prague, Czech Republic, 13–17 July 2019; pp. 263–264. [Google Scholar]
- Freund, Y.; Schapire, R.E. A decision-theoretic generalization of on-line learning and an application to boosting. J. Comput. Syst. Sci. 1997, 55, 119–139. [Google Scholar] [CrossRef]
- Arora, S.; Hazan, E.; Kale, S. The Multiplicative Weights Update Method: a Meta-Algorithm and Applications. Theory Comput. 2012, 8, 121–164. [Google Scholar] [CrossRef]
- Biguri, A.; Dosanjh, M.; Hancock, S.; Soleimani, M. TIGRE: a MATLAB-GPU toolbox for CBCT image reconstruction. Biomed. Phys. Eng. Exp. 2016, 2, 055010. [Google Scholar] [CrossRef]
- Andersen, A.H.; Kak, A.C. Simultaneous algebraic reconstruction technique (SART): a superior implementation of the ART algorithm. Ultrason. Imaging 1984, 6, 81–94. [Google Scholar] [CrossRef]
- Arlot, S.; Celisse, A. A survey of cross-validation procedures for model selection. Stat. Surv. 2010, 4, 40–79. [Google Scholar] [CrossRef]
- Tibshirani, R. Regression shrinkage and selection via the lasso. J. R. Stat. Soc. Ser. B (Methodol.) 1996, 58, 267–288. [Google Scholar] [CrossRef]
- Chretien, S.; Gibberd, A.; Roy, S. Hedging hyperparameter selection for basis pursuit. arXiv preprint 2018, arXiv:1805.01870. [Google Scholar]
- Heidrich-Meisner, V.; Igel, C. Hoeffding and Bernstein races for selecting policies in evolutionary direct policy search. In Proceedings of the 26th Annual International Conference on Machine Learning, Montreal, QC, Canada, 14–18 June 2009; pp. 401–408. [Google Scholar]
- Segars, W.; Sturgeon, G.; Mendonca, S.; Grimes, J.; Tsui, B.M. 4D XCAT phantom for multimodality imaging research. Med. Phys. 2010, 37, 4902–4915. [Google Scholar] [CrossRef]
- Xu, J.; Tsui, B.M. Electronic noise modeling in statistical iterative reconstruction. IEEE Trans. Image Process. 2009, 18, 1228–1238. [Google Scholar]
- Ma, J.; Liang, Z.; Fan, Y.; Liu, Y.; Huang, J.; Chen, W.; Lu, H. Variance analysis of x-ray CT sinograms in the presence of electronic noise background. Med. Phys. 2012, 39, 4051–4065. [Google Scholar] [CrossRef] [PubMed]
- Wang, Z.; Bovik, A.C. A universal image quality index. IEEE Signal Process. Lett. 2002, 9, 81–84. [Google Scholar] [CrossRef]
- Zhang, H.; Ouyang, L.; Huang, J.; Ma, J.; Chen, W.; Wang, J. Few-view cone-beam CT reconstruction with deformed prior image. Med. Phys. 2014, 41, 121905. [Google Scholar] [CrossRef] [PubMed]
Hyper-Parameters | Values |
---|---|
0,50,70,100,200,500, | |
2,,, | |
2,4,6,8,10,12,14,16,18,20,22,24,26,28,30 | |
1 | |
0.99 | |
0.0212 |
Approaches | |||||
---|---|---|---|---|---|
Hedge-based approach | 0 | 10 | 1 | 0.99 | 0.0212 |
Cross-Validation approach | 0 | 8 | 1 | 0.99 | 0.0212 |
Approaches | Relative Errors (%) | UQI | Computational Time |
---|---|---|---|
(h) | |||
Hedge-based | 4.9349 | 0.9972 | 16.12 |
Cross-validation | 5.2597 | 0.9970 | 47.15 |
Parametrisation Details | Male Phantom | Female Phantom |
---|---|---|
motion option | beating heart only | respiratory only |
length of beating heart cycle | 1 sec | 5 secs |
starting phase of the heart | 0.0 | 0.4 |
wall thickness for the left | ||
ventricle(LV) | non-uniform | uniform |
LV end-systolic volume | 0.0 | 0.5 |
start phase of the respiratory | 0.0 | 0.4 |
anteroposterior diameter | ||
of the ribcage, body and lungs | 0.5 | 1.2 |
heart’s lateral motion | ||
during breathing | 0.0 | 0.5 |
heart’s up/down motion | ||
during breathing | 2.0 | 3.0 |
breast type | prone | supine |
factor to compress breast | half compression | no compression |
thickness of sternum | 0.4 | 0.6 |
thickness of scapula | 0.35 | 0.55 |
thickness of ribs | 0.3 | 0.5 |
thickness of backbone | 0.4 | 0.6 |
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Share and Cite
Chrétien, S.; Lohvithee, M.; Sun, W.; Soleimani, M. Efficient Hyper-Parameter Selection in Total Variation-Penalised XCT Reconstruction Using Freund and Shapire’s Hedge Approach. Mathematics 2020, 8, 493. https://doi.org/10.3390/math8040493
Chrétien S, Lohvithee M, Sun W, Soleimani M. Efficient Hyper-Parameter Selection in Total Variation-Penalised XCT Reconstruction Using Freund and Shapire’s Hedge Approach. Mathematics. 2020; 8(4):493. https://doi.org/10.3390/math8040493
Chicago/Turabian StyleChrétien, Stéphane, Manasavee Lohvithee, Wenjuan Sun, and Manuchehr Soleimani. 2020. "Efficient Hyper-Parameter Selection in Total Variation-Penalised XCT Reconstruction Using Freund and Shapire’s Hedge Approach" Mathematics 8, no. 4: 493. https://doi.org/10.3390/math8040493
APA StyleChrétien, S., Lohvithee, M., Sun, W., & Soleimani, M. (2020). Efficient Hyper-Parameter Selection in Total Variation-Penalised XCT Reconstruction Using Freund and Shapire’s Hedge Approach. Mathematics, 8(4), 493. https://doi.org/10.3390/math8040493