The denoising of Monte Carlo dose distributions using convolution superposition calculations

I. El Naqa, J. Cui, P. Lindsay, G. Olivera, J. O. Deasy

Research output: Contribution to journalArticle

Abstract

Monte Carlo (MC) dose calculations can be accurate but are also computationally intensive. In contrast, convolution superposition (CS) offers faster and smoother results but by making approximations. We investigated MC denoising techniques, which use available convolution superposition results and new noise filtering methods to guide and accelerate MC calculations. Two main approaches were developed to combine CS information with MC denoising. In the first approach, the denoising result is iteratively updated by adding the denoised residual difference between the result and the MC image. Multi-scale methods were used (wavelets or contourlets) for denoising the residual. The iterations are initialized by the CS data. In the second approach, we used a frequency splitting technique by quadrature filtering to combine low frequency components derived from MC simulations with high frequency components derived from CS components. The rationale is to take the scattering tails as well as dose levels in the high-dose region from the MC calculations, which presumably more accurately incorporates scatter; high-frequency details are taken from CS calculations. 3D Butterworth filters were used to design the quadrature filters. The methods were demonstrated using anonymized clinical lung and head and neck cases. The MC dose distributions were calculated by the open-source dose planning method MC code with varying noise levels. Our results indicate that the frequency-splitting technique for incorporating CS-guided MC denoising is promising in terms of computational efficiency and noise reduction.

Original languageEnglish (US)
JournalPhysics in Medicine and Biology
Volume52
Issue number17
DOIs
StatePublished - Sep 7 2007
Externally publishedYes

Fingerprint

Convolution
convolution integrals
Noise
dosage
Monte Carlo Method
quadratures
Neck
Head
Lung
Butterworth filters
filters
Computational efficiency
Noise abatement
noise reduction
lungs
iteration
Monte Carlo method
planning
Monte Carlo methods
Scattering

ASJC Scopus subject areas

  • Biomedical Engineering
  • Physics and Astronomy (miscellaneous)
  • Radiology Nuclear Medicine and imaging
  • Radiological and Ultrasound Technology

Cite this

The denoising of Monte Carlo dose distributions using convolution superposition calculations. / Naqa, I. El; Cui, J.; Lindsay, P.; Olivera, G.; Deasy, J. O.

In: Physics in Medicine and Biology, Vol. 52, No. 17, 07.09.2007.

Research output: Contribution to journalArticle

Naqa, I. El ; Cui, J. ; Lindsay, P. ; Olivera, G. ; Deasy, J. O. / The denoising of Monte Carlo dose distributions using convolution superposition calculations. In: Physics in Medicine and Biology. 2007 ; Vol. 52, No. 17.
@article{926df732aebe4959adca13ab280321d0,
title = "The denoising of Monte Carlo dose distributions using convolution superposition calculations",
abstract = "Monte Carlo (MC) dose calculations can be accurate but are also computationally intensive. In contrast, convolution superposition (CS) offers faster and smoother results but by making approximations. We investigated MC denoising techniques, which use available convolution superposition results and new noise filtering methods to guide and accelerate MC calculations. Two main approaches were developed to combine CS information with MC denoising. In the first approach, the denoising result is iteratively updated by adding the denoised residual difference between the result and the MC image. Multi-scale methods were used (wavelets or contourlets) for denoising the residual. The iterations are initialized by the CS data. In the second approach, we used a frequency splitting technique by quadrature filtering to combine low frequency components derived from MC simulations with high frequency components derived from CS components. The rationale is to take the scattering tails as well as dose levels in the high-dose region from the MC calculations, which presumably more accurately incorporates scatter; high-frequency details are taken from CS calculations. 3D Butterworth filters were used to design the quadrature filters. The methods were demonstrated using anonymized clinical lung and head and neck cases. The MC dose distributions were calculated by the open-source dose planning method MC code with varying noise levels. Our results indicate that the frequency-splitting technique for incorporating CS-guided MC denoising is promising in terms of computational efficiency and noise reduction.",
author = "Naqa, {I. El} and J. Cui and P. Lindsay and G. Olivera and Deasy, {J. O.}",
year = "2007",
month = "9",
day = "7",
doi = "10.1088/0031-9155/52/17/N02",
language = "English (US)",
volume = "52",
journal = "Physics in Medicine and Biology",
issn = "0031-9155",
publisher = "IOP Publishing Ltd.",
number = "17",

}

TY - JOUR

T1 - The denoising of Monte Carlo dose distributions using convolution superposition calculations

AU - Naqa, I. El

AU - Cui, J.

AU - Lindsay, P.

AU - Olivera, G.

AU - Deasy, J. O.

PY - 2007/9/7

Y1 - 2007/9/7

N2 - Monte Carlo (MC) dose calculations can be accurate but are also computationally intensive. In contrast, convolution superposition (CS) offers faster and smoother results but by making approximations. We investigated MC denoising techniques, which use available convolution superposition results and new noise filtering methods to guide and accelerate MC calculations. Two main approaches were developed to combine CS information with MC denoising. In the first approach, the denoising result is iteratively updated by adding the denoised residual difference between the result and the MC image. Multi-scale methods were used (wavelets or contourlets) for denoising the residual. The iterations are initialized by the CS data. In the second approach, we used a frequency splitting technique by quadrature filtering to combine low frequency components derived from MC simulations with high frequency components derived from CS components. The rationale is to take the scattering tails as well as dose levels in the high-dose region from the MC calculations, which presumably more accurately incorporates scatter; high-frequency details are taken from CS calculations. 3D Butterworth filters were used to design the quadrature filters. The methods were demonstrated using anonymized clinical lung and head and neck cases. The MC dose distributions were calculated by the open-source dose planning method MC code with varying noise levels. Our results indicate that the frequency-splitting technique for incorporating CS-guided MC denoising is promising in terms of computational efficiency and noise reduction.

AB - Monte Carlo (MC) dose calculations can be accurate but are also computationally intensive. In contrast, convolution superposition (CS) offers faster and smoother results but by making approximations. We investigated MC denoising techniques, which use available convolution superposition results and new noise filtering methods to guide and accelerate MC calculations. Two main approaches were developed to combine CS information with MC denoising. In the first approach, the denoising result is iteratively updated by adding the denoised residual difference between the result and the MC image. Multi-scale methods were used (wavelets or contourlets) for denoising the residual. The iterations are initialized by the CS data. In the second approach, we used a frequency splitting technique by quadrature filtering to combine low frequency components derived from MC simulations with high frequency components derived from CS components. The rationale is to take the scattering tails as well as dose levels in the high-dose region from the MC calculations, which presumably more accurately incorporates scatter; high-frequency details are taken from CS calculations. 3D Butterworth filters were used to design the quadrature filters. The methods were demonstrated using anonymized clinical lung and head and neck cases. The MC dose distributions were calculated by the open-source dose planning method MC code with varying noise levels. Our results indicate that the frequency-splitting technique for incorporating CS-guided MC denoising is promising in terms of computational efficiency and noise reduction.

UR - http://www.scopus.com/inward/record.url?scp=36048940745&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=36048940745&partnerID=8YFLogxK

U2 - 10.1088/0031-9155/52/17/N02

DO - 10.1088/0031-9155/52/17/N02

M3 - Article

C2 - 17762073

AN - SCOPUS:36048940745

VL - 52

JO - Physics in Medicine and Biology

JF - Physics in Medicine and Biology

SN - 0031-9155

IS - 17

ER -