Accelerated Prompt Gamma estimation for clinical Proton Therapy simulations

HAL Id: hal-01276369

https://hal.archives-ouvertes.fr/hal-01276369

Submitted on 14 Feb 2017

HAL is a multi-disciplinary open access

archive for the deposit and dissemination of sci-entific research documents, whether they are pub-lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Accelerated Prompt Gamma estimation for clinical Proton Therapy simulations

B. Huisman, Jean Michel Létang, E. Testa, D. Sarrut

To cite this version:

Accelerated Prompt Gamma estimation for

clinical Proton Therapy simulations

B.F.B. Huisman

, J.M. Létang

, É. Testa

, D. Sarrut

1 _{CREATIS, Université de Lyon; CNRS UMR5220; INSERM U1044; INSA-Lyon; Université Lyon 1; Centre Léon Bérard, Lyon, France}

2 _{IPNL, Université de Lyon; CNRS/IN2P3 UMR5822; Université Lyon 1 Lyon, France}

brent.huisman@creatis.insa-lyon.fr

1. P

URPOSE

There is interest in the particle therapy

community to use prompt gammas (PG), a natural byproduct of particle treatment, for range verification and eventually dose

control (Knopf et al. 2015). However, PG

production is a rare process and therefore estimating PGs exiting a patient during a proton treatment plan executed by a Monte

Carlo simulation (MC) converges slowly.

Primaries PGs Exiting patient Solid angle detector Post-collimator Detector Efficiency ? Reconstruction Eff. ? 103 104 105 106 107 108 109 Coun ts Protons Prompt Gammas

We present a generic PG yield estimator, drop-in usable with any geometry and beam configuration. We show a gain of three orders of magnitude compared to analog MC. We analyze the depth profile and the PG energy spectrum of a simple phantom and a clinical head and neck CT image.

2. C

ONCEPT

1 2 3

1. Regular Monte Carlo tracking

A regular MC simulation propagates particles throughout geometry. The propagation is broken up into steps, at which point the engine compiles a list of all possible futures, weights them, and using a random number selects the actual future.

2. At each step: Prompt Gamma production probability

Parallel to executing this conventional tracking, we may request and store the PG production probabilities. At each step, as function of PG energy, a production probability spectrum is stored at the current voxel.

3. Limited MC to touch all relevant voxels

By propagating a number of primary protons in this way, we obtain probabilities in all the voxels that a beam may touch. We need a minimum number of primaries, since we can only request PG probabilities in the voxels the primary passes

through. However, we require fewer primary

propagations with respect to a fully analog MC.

A

CKNOWLEDGMENTS

This work was partly supported by Labex PRIMES ANR-11-LABX-0063, t-Gate ANR-14-CE23-0008, France Hadron ANR-11-INBS-0007 and LYric INCa-DGOS-4664.

3. M

ETHOD

A voxelized Prompt-Gamma Track Length

Estimator (Kanawati et al. 2015) simulation is broken up into two stages. A PGdb (Stage 0) is

presupposed, computed once and reused. It

contains an estimate of the effective linear PG

production coefficient ΓΓΓ_Z modulo the density

ρZ , per element (k). At the start of Stage 1,

the coefficients are computed for the materials found in the phantom (eq. 1).

ΓΓΓm(E ) = ρm_v k_mv X k=1 ωk ΓΓΓZ_k (E ) ρZ_k (1) b SbSSbi (v) = ΓΓΓm_v (Eg )Lg (Eg , v) (2) Per step, per voxel v in the PGyd, alongside executing the analog MC processes, we compute

and add the product of the step length L_g

and ΓΓΓ_mv , with m_v the material at voxel v and

g the proton energy bin (eq. 2). Put into

words, we compute the PG yield probability energy spectrum at every step, and add it to any pre-existing spectrum in the current voxel

v. The PGyd computed in stage 1 is used as

a PG production source in Stage 2. If the user

is interested in the PG signal of 1011 protons,

the PGyd can be requested to give the expected output for that number of protons. Each PG is then propagated through the geometry and into the detector with regular analog MC processes.

4. R

ESULT

S

IMPLE PHANTOM

0 50 100 150 200 0.0 0.5 1.0 1.5 2.0 2.5 In tegrated Yield [PG/proton/v oxel] ×10−3 1 2 3 4 5 6 7 8 0.0 0.5 1.0 1.5 2.0 2.5 ×10−3 ₁₀3 _primaries 104 primaries 105 primaries 106 primaries Reference 0 50 100 150 200 −3 −2 −1 0 1 2 3 In tegrated Rel. Diff.[%] 1 2 3 4 5 6 7 8 −3 −2 −1 0 1 2 3 0 50 100 150 200 Depth [mm] −6 −4 −2 0 2 4 6 V oxels b eam path Rel. Diff.[%] 1 2 3 4 5 6 7 8 PG energy [MeV] −6 −4 −2 0 2 4 6 102 103 104

Gain factor w.r.t. Reference

0.0 0.2 0.4 0.6 0.8 1.0 Num b er of vo xels (scaled)

vpgTLE gain distribution Median gain: 1.40 _{× 10}3 103 primaries Min: 6.30_{× 10}1 Max: 4.64_{× 10}4 104 primaries Min: 6.19× 101 Max: 3.73× 104 105 primaries Min: 9.03_{× 10}1 Max: 5.21_{× 10}4 106 primaries Min: 8.63× 101 Max: 3.21× 104 101 102 103 104 105 106 Runtime t [s] 0 2 4 6 8 10 12 Relativ e Uncertain ty [%]

Median relative uncertainty Gain: 1.55 _{× 10}3 vpgTLE, Fit: 2.3×10−1_√ t Analog, Fit: 8.9×100_√ t

5. R

ESULT

C

LINICAL PHANTOM

0 20 40 60 80 100 120 140 160 0.0 0.5 1.0 1.5 2.0 In tegrated Yield [PG/proton/bin] ×10−3 1 2 3 4 5 6 7 8 0.0 0.5 1.0 1.5 2.0 ×10−3 103 primaries 104 primaries 105 primaries 106 primaries Reference 0 20 40 60 80 100 120 140 160 Depth [mm] −3 −2 −1 0 1 2 3 In tegrated Rel. Diff.[%] 1 2 3 4 5 6 7 8 PG energy [MeV] −3 −2 −1 0 1 2 3 102 103 104

Gain factor w.r.t. Reference

0.0 0.2 0.4 0.6 0.8 1.0 Num b er of vo xels (scaled)

vpgTLE gain distribution Median gain: 9.98 _{× 10}2 103 primaries Min: 0 Max: 2.76× 105 104 primaries Min: 3.85_{× 10}1 Max: 3.29× 104 105 primaries Min: 4.70× 101 Max: 4.88_{× 10}4 106 primaries Min: 2.70_{× 10}1 Max: 8.96× 104 102 103 104 105 106 107 Runtime t [s] 0 10 20 30 40 50 60 70 Relativ e Uncertain ty [%]

Median relative uncertainty Gain: 1.03 _{× 10}3 vpgTLE, Fit: 2.5_√×100 t Analog, Fit: 7.9_√×101 t

6. C

ONCLUSION

vpgTLE is a generic drop-in alternative for computing the expected PG output in voxelized

geometries. The method reaches a global

gain factor of 101010333 for a clinical CT image and treatment plan with respect to analog MC. A median convergence of 2% for the most distal energy layer is reached in approximately four

hours on a single core, with the output stabilized

to within 10−4 of an analog reference simulation,

when the PG yield along proton range and PG

spectrum are considered. Those interested in

developing and simulating PG detection devices, as well as clinicians studying complex clinical cases, may benefit from the precision and accuracy of vpgTLE simulations not offered by analytic algorithms.

The vpgTLE method is open source, fully integrated and available in the next Gate release. This study has been submitted to Physics in

Medicine and Biology.

R

EFERENCES

Knopf et al. (2015) Phys. Med. Biol.