# Mesh Optimization for Monte Carlo-Based Optical Tomography

^{*}

## Abstract

**:**

## 1. Introduction

^{6}–10

^{9}) per simulated optode, depending on the data type, in preclinical settings [14]. As the accuracy of the MC-based Jacobian is dependent on the local statistics of the forward problem, coarsening or refining the mesh requires recomputing the Jacobian, possibly with a greater photon packet number to satisfy smaller discretizations. The iterative nature of adaptive mesh techniques may then lead to hours of computations, even in a massively parallel environment. Herein, we investigate the application of fast and efficient forward mesh optimization approaches for time resolved MC-based FMT. We propose a mesh optimization methodology in which the initial mesh and MC forward models are analytically rescaled at each iteration, allowing for fast computation without loss of accuracy.

## 2. Methods

#### 2.1. Optical Inverse Problem

#### 2.2. Mesh-Based Monte Carlo

#### 2.3. Mesh Adaptation

#### 2.4. Jacobian Rescaling

**Figure 3.**The possible cases in the output node positioning. The upper row corresponds to the input discretization and lower row to the output mesh.

#### 2.5. In Silico Model

**Figure 4.**(

**a**) The digimouse model with slice highlighted next to (

**b**) the mesh section used with positions of sources (solid black) and detectors (solid grey). The slice is 4 mm thick.

## 3. Results and Discussion

#### 3.1. Size Factors

Size Factor | Iterations | Nodes | Elements | Max. Elem. Vol. (mm^{3}) |
---|---|---|---|---|

Initial | -- | 2396 | 7574 | 0.08 |

1.15 | 17 | 1332 | 3904 | 0.51 |

1.25 | 21 | 777 | 2200 | 1.49 |

1.35 | 20 | 612 | 1615 | 2.97 |

**Figure 5.**The mesh under different maximum size factors after (

**a**) one iteration and (

**b**) seventeen iterations, the greatest iteration all meshes achieve before convergence. Maximum size factors are provided as subtitles (color bar corresponds to volume of elements in mm

^{3}).

#### 3.2. Forward Model-Based Solution Fields

Field | Iterations | Nodes | Elements | Max. Elem. Vol. (mm^{3}) |
---|---|---|---|---|

∑Jacobian | 21 | 777 | 2200 | 1.49 |

Log(∑Jacobian) | 12 | 1487 | 4335 | 0.25 |

Normalized ∑J | 21 | 1053 | 3074 | 0.48 |

Curvature | 15 | 447 | 1105 | 1.55 |

Log(Curv.) | 20 | 668 | 1651 | 1.37 |

^{3}, the smallest maximum volumes achieved at convergence. This allows for comparison of all solution fields as summarized in Table 3.

Field | Iterations | Nodes | Elements | Max. Elem. Vol. (mm^{3}) |
---|---|---|---|---|

∑Jacobian | 2 | 1614 | 4872 | 0.26 |

Log(∑Jacobian) | 5 | 1611 | 4792 | 0.25 |

Normalized ∑J | 5 | 1457 | 4334 | 0.25 |

Curvature | 2 | 1136 | 3304 | 0.31 |

Log(Curv.) | 2 | 1608 | 4846 | 0.25 |

**Figure 6.**Upper row: mesh at convergence (

**a**) ${{\displaystyle \sum}}^{\text{}}Jacobian$, (

**b**) $log\left({{\displaystyle \sum}}^{\text{}}Jacobian\right)$, (

**c**) $normalized{{\displaystyle \sum}}^{\text{}}J$; (

**d**) $u$ and (

**e**) $Log\left(u\right)$; lower row: mesh at 0.25 mm

^{3}stopping criterion for the respective corresponding input fields.

#### 3.3. Geometry-Based Solution Fields

Field | Iterations | Nodes | Elements | Max. Elem. Vol. (mm^{3}) |
---|---|---|---|---|

∑Jacobian | 21 | 777 | 2200 | 1.49 |

Dist. | 11 | 1675 | 5166 | 0.28 |

Att. | 17 | 699 | 1999 | 1.80 |

#### 3.4. Jacobian Accuracy and Computational Efficiency

^{9}photons) prior to adaptation and from the analytically rescaled Jacobian after adaptation (10

^{9}photons). Moreover, new mMC simulations were computed to obtain a mMC Jacobian on the new adapted mesh. An example of TPSFs produced for one specific source detector pair is provided in Figure 8a) (central pair). The TPSFs simulated under the original and mesh adaptation conditions are matching remarkably. Furthermore, the error in the TPSF from the rescaled Jacobian is similar to the error in the TPSF from the recomputed Jacobian. Throughout all time gates that are typically employed to cast the inverse problem, the maximum error is less than 1.5% for late gates and below 0.5% for early gates (Figure 8b). This error level is similar for all 49 simulated optode combinations. These results indicate that the mesh adaptation and analytical rescaling methodology described above lead to optimized non-homogenous discretization that does not affect light propagation accuracy, even in the most challenging cases, i.e., early gates.

**Figure 8.**Temporal Point Spread Function (TPSF) of Jacobians rescaled to a new mesh and the associated error at each gate.

**Figure 9.**(

**a**) Elements of volume employed to compute the error $e\left(r\right)$ (red label); (

**b**) Computation time for one forward simulation versus number of photons.

^{10}photons per optode. The error was computed for the Jacobian at the time corresponding to the 25% rising gate of the TPSF. A summary of the $e\left(r\right)$ calculated for this configuration and for different photon packets is provided in Table 5.

**Table 5.**Errors in the forward model central nodes before and after ${{\displaystyle \sum}}^{\text{}}Jacobian$ mesh optimization.

Photons | Initial TG | Final TG ∑Jacobian | Final TG $Att.$ |
---|---|---|---|

10^{9} | 10.84% | 11.71% | 13.99% |

10^{8} | 36.83% | 31.39% | 35.55% |

10^{7} | 61.01% | 60.13% | 58.57% |

10^{6} | 86.71% | 90.92% | 151.43% |

^{6}photons, $e\left(r\right)$ was similar between initial and rescaled meshes/Jacobians. Note that repeating 10

^{10}photons simulations on the initial mesh led to $e\left(r\right)~2\%$. Overall, 10

^{9}produced the least $e\left(r\right)$ with both field metric performing well, though they had 32% (∑Jacobian) and 34% (Attenuation Field) less nodes. The reduction in nodes and elements is mainly achieved in the central part of the mesh where the elements of volume are less visited by photons. The coarsening of the mesh in this region leads to improved statistics as seen in $e\left(r\right)$ for early gates (note that $e\left(r\right)$ is the average error as computed by the mean of the error at each node). In the case of 10

^{8}photons, which is the typical number of photons used successfully for preclinical studies [14], the coarsening leads to a 14% reduction in error $e\left(r\right)$. This suggests that mesh optimization as described herein can lead to reduction of the size of photon packets simulated and reduction in nodes/elements of volume for a more tractable inverse problem.

## 4. Conclusions

## Acknowledgments

## Author Contributions

## Conflict of Interest

## References

- Darne, C.; Lu, Y.; Sevick-Muraca, E.M. Small animal fluorescence and bioluminescence tomography: a review of approaches, algorithms and technology update. Phys. Med. Biol.
**2014**, 59, R1–R64. [Google Scholar] [CrossRef] [PubMed] - Chen, Y.; Mu, C.; Intes, X.; Blessington, D.; Chance, B. Near-infrared phase cancellation instrument for fast and accurate localization of fluorescent heterogeneity. Rev. Sci. Instrum.
**2003**, 74, 3466–3473. [Google Scholar] [CrossRef] - Venugopal, V.; Chen, J.; Barrosso, M.; Intes, X. Quantitative tomographic imaging of intermolecular FRET in small animals. Biomed. Opt. Express
**2012**, 3, 3161–3175. [Google Scholar] [CrossRef] [PubMed] - Arridge, S.R.; Hebden, J.C. Optical imaging in medicine: II. Modelling and reconstruction. Phys. Med. Biol.
**1997**, 42, 841–853. [Google Scholar] - Ren, K.; Bal, G.; Hielscher, A.H. Transport- and diffusion-based optical tomography in small domains: A comparative study. Appl. Opt.
**2007**, 46, 6669–6679. [Google Scholar] [CrossRef] [PubMed] - Dehghani, H.; Eames, M.E.; Yalavarthy, P.K.; Davis, S.C.; Srinivasan, S.; Carpenter, C.M.; Pogue, B.W.; Paulsen, K.D. Near Infrared optical tomography using NIRFAST: Algorithm for numerical model and image reconstruction. Commu. Numer. Methods Eng.
**2008**, 25, 711–732. [Google Scholar] [CrossRef] - Yoo, K.M.; Liu, F.; Alfano, R.R. When does diffusion approximation fail to describe photon transport in random media. Phys. Rev. Lett.
**1990**, 64, 2647–2650. [Google Scholar] [CrossRef] [PubMed] - Hielscher, A.H.; Alcouffe, R.E.; Barbour, R.L. Comparison of finite-difference transport and diffusion calculations for photon migration in homogeneous and heterogeneous tissues. Phys. Med. Biol.
**1998**, 43, 1285–1302. [Google Scholar] [CrossRef] [PubMed] - Arridge, S.R.; Dehghani, H.; Schweiger, M.; Okada, A.E. The finite element model for the propagation of light in scattering media: a direct method for domains with nonscattering regions. Med. Phys.
**2000**, 27, 252–264. [Google Scholar] [CrossRef] [PubMed] - Sakami, M.; Mitra, K.; Vo-Dinh, T. Analysis of short-pulse laser photon transport through tissues for optical tomography. Opt. Lett.
**2002**, 27, 336–338. [Google Scholar] [CrossRef] - Zhu, C.; Liu, Q. Review of Monte Carlo modeling of light transport in tissues. J. Biomed. Opt.
**2013**, 18. [Google Scholar] [CrossRef] - Chen, J.; Intes, X. Time-gated perturbation Monte Carlo for whole body functional imaging in small animals. Opt. Express
**2009**, 17, 19566–19579. [Google Scholar] [CrossRef] [PubMed] - Chen, J.; Venugopal, V.; Intes, X. Monte Carlo based method for fluorescence tomographic imaging with lifetime multiplexing using time gates. Biomed. Opt. Express
**2011**, 2, 871–886. [Google Scholar] [CrossRef] [PubMed] - Chen, J.; Intes, X. Comparison of Monte Carlo Methods for Fluorescence Molecular Tomography–Computational Efficiency. Med. Phys.
**2011**, 38, 5788–5798. [Google Scholar] [CrossRef] [PubMed] - Fang, Q.; Kaeli, D.R. Accelerating mesh-based Monte Carlo method on modern CPU architectures. Biomed. Opt. Express
**2012**, 3, 3223–3230. [Google Scholar] [CrossRef] [PubMed] - Valim, N.; Brock, J.; Niedre, M.J. Experimental measurement of time-dependent photon scatter for diffuse optical tomography. J. Biomed. Opt.
**2010**, 5. [Google Scholar] [CrossRef] - Boas, D.A.; Culver, J.P.; Stott, J.J.; Dunn, A.K. Three dimensional Monte Carlo code for photon migration through complex heterogeneous media including the adult human head. Opt. Express
**2002**, 10, 159–170. [Google Scholar] [CrossRef] [PubMed] - Ren, K.; Abdoulaev, G.S.; Bal, G.; Hielscher, A.H. Algorithm for solving the equation of radiative transfer in the frequency domain. Opt. Lett.
**2004**, 29, 578–580. [Google Scholar] [CrossRef] [PubMed] - Hayakawa, C.K.; Spanier, J. Perturbation Monte Carlo methods to solve inverse photon migration problems in heterogeneous tissues. Opt. Lett.
**2001**, 26, 1335–1337. [Google Scholar] [CrossRef] [PubMed] - Kumar, Y.P.; Vasu, R.M. Reconstruction of optical properties of low-scattering tissue using derivative estimated through perturbation Monte-Carlo method. J. Biomed. Opt.
**2004**, 9, 1002–1012. [Google Scholar] [CrossRef] [PubMed] - Kirkby, D.R.; Delpy, D.T. Parallel operation of Monte Carlo simulations on a diverse network of computers. Phys. Med. Biol.
**1997**, 42, 1203–1208. [Google Scholar] [CrossRef] [PubMed] - Alerstam, E.; Svensson, T.; Andersson-Engels, S. Parallel computing with graphics processing units for high-speed Monte Carlo simulation of photon migration. J. Biomed. Opt.
**2008**. [Google Scholar] [CrossRef] - Venugopal, V.; Chen, J.; Lesage, F.; Intes, X. Full-field time-resolved fluorescence tomography of small animals. Opt. Lett.
**2010**, 35, 3189–3191. [Google Scholar] [CrossRef] [PubMed] - Fang, F. Mesh-based Monte Carlo method using fast ray-tracing in Plücker coordinates. Biomed. Opt. Express
**2010**, 1, 165–175. [Google Scholar] [CrossRef] [PubMed] - Shen, H.; Wang, G. A study on tetrahedron-based inhomogeneous Monte Carlo optical simulation. Biomed. Opt. Express
**2011**, 2, 44–57. [Google Scholar] [CrossRef] - Chen, J.; Fang, Q.; Intes, X. Mesh-based Monte Carlo method in time-domain wide-field fluorescence molecular tomography. J. Biomed. Opt.
**2012**, 17. [Google Scholar] [CrossRef] - Pimpalkhare, M.; Chen, J.; Venugopal, V.; Intes, X. Ex vivo fluorescence molecular tomography of the spine. J. Biomed. Imaging
**2012**. [Google Scholar] [CrossRef] - Zhao, L.; Yang, H.; Cong, W.; Wang, G.; Intes, X. Lp Regularization for Early Time-Gate Fluorescence Molecular Tomography. Opt. Lett.
**2014**, 39, 4156–4159. [Google Scholar] [CrossRef] [PubMed] - Chen, J.; Venugopal, V.; Lesage, F.; Intes, X. Time Resolved Diffuse Optical Tomography with patterned light illumination and detection. Opt. Lett.
**2010**, 35, 2121–2123. [Google Scholar] - Pian, P.; Yao, R.; Zhao, L.; Intes, X. Hyperspectral Time-Resolved Wide-Field Fluorescence Molecular Tomography based on Structured Light and Single Pixel-Detection. Opt. Lett.
**2015**, 40, 431–434. [Google Scholar] [CrossRef] - Bangerth, W. A framework for the adaptive finite element solution of large-scale inverse problems. SIAM J. Sci. Comput.
**2008**, 30, 2965–2989. [Google Scholar] [CrossRef] - Arridge, S.R.; Schotland, J.C. Optical tomography: Forward and inverse problems. Inverse Probl.
**2009**, 25, 123010. [Google Scholar] [CrossRef] - Thomas, D.M.; Yalavarthy, P.; Karkalla, D.; Natarajan, V. Mesh simplification based on edge collapsing could improve computational efficiency in Near infrared optical tomography. IEEE J. Sel. Top. Quantum Electron.
**2012**, 18, 1493–1501. [Google Scholar] - Gu, X.; Xu, Y.; Jiang, H. Mesh-based enhancement scheme in diffuse optical tomography. Med. Phys.
**2003**, 30, 861–869. [Google Scholar] [CrossRef] [PubMed] - Joshi, A.; Bangerth, W.; Sevick-Muraca, E.M. Adaptive finite element based tomography for fluorescence optical imaging in tissue. Opt. Express
**2004**, 12, 5402–5417. [Google Scholar] [CrossRef] [PubMed] - Soloviev, V.Y.; Krasnosselskaia, L.V. Dynamically adaptive mesh refinement technique for image reconstruction in optical tomography. Appl. Opt.
**2006**, 45, 2828–2837. [Google Scholar] [CrossRef] [PubMed] - Guven, M.; Giladi, E.; Yazici, B.; Kwon, K.; Intes, X. Effect of discretization error and adaptive mesh generation in diffuse optical absorption imaging: Part I. Inverse Probl.
**2007**, 23, 1115–1133. [Google Scholar] [CrossRef] - Lee, J.H.; Joshi, A.; Sevick-Muraca, E.M. Fully adaptive finite element based tomography using tetrahedral dual-meshing for fluorescence enhanced optical imaging in tissue. Opt. Express
**2007**, 15, 6955–6975. [Google Scholar] [CrossRef] [PubMed] - Guo, H.; He, X.; Yu, J.; Cheng, J.; Pu, X. Adaptive hp finite element method for fluorescence molecular tomography with simplified spherical harmonics approximation. J. Innov. Opt. Health Sci.
**2014**, 7. [Google Scholar] [CrossRef] - Song, X.; Wang, D.; Chen, N.; Bai, J.; Wang, H. Reconstruction for free space fluorescence tomography using a novel hybrid adaptive finite element algorithm. Opt. Express
**2007**, 15, 18300–18317. [Google Scholar] [CrossRef] [PubMed] - Zhou, L.; Yazici, B.; Ale, A.B.F.; Ntziachristos, V. Performance evaluation of adaptive meshing algorithms for fluorescence diffuse optical tomogrpahy using experimental data. Opt. Lett.
**2010**, 15, 3727–3729. [Google Scholar] [CrossRef] - Prakash, J.; Chandrasekharan, V.; Upendra, V.; Yalavarthy, P.K. Accelerating frequency-domain diffuse optical tomographic image reconstruction using graphics processing units. J. Biopmed. Opt.
**2010**, 15. [Google Scholar] [CrossRef] - Li, X. Mesh modification procedures for general 3D non-manifold domains. Ph.D. Dissertation, Department Mechanical Engineering, Rensselaer Polytechnic Institute, Troy, NY, USA, 2003. [Google Scholar]
- Venugopal, V.; Chen, J.; Intes, X. Robust imaging strategies in time-resolved optical tomography. Proc. SPIE
**2013**. [Google Scholar] [CrossRef] - Holt, R.W.; Leblond, F.L.; Pogue, B.W. Methodology to optimize detector geometry in fluorescence tomography of the tissue using minimized curvature of the summed diffuse sensitivity projections. J. Opt. Soc. Am. A Opt. Image Sci. Vis.
**2013**, 30, 1613–1619. [Google Scholar]

© 2015 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 license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Edmans, A.; Intes, X. Mesh Optimization for Monte Carlo-Based Optical Tomography. *Photonics* **2015**, *2*, 375-391.
https://doi.org/10.3390/photonics2020375

**AMA Style**

Edmans A, Intes X. Mesh Optimization for Monte Carlo-Based Optical Tomography. *Photonics*. 2015; 2(2):375-391.
https://doi.org/10.3390/photonics2020375

**Chicago/Turabian Style**

Edmans, Andrew, and Xavier Intes. 2015. "Mesh Optimization for Monte Carlo-Based Optical Tomography" *Photonics* 2, no. 2: 375-391.
https://doi.org/10.3390/photonics2020375