The potential of dual-energy CT to reduce proton beam range uncertainties

Bär, Esther; Lalonde, Arthur; Royle, G.; Lu, Hsiao‐Ming; Bouchard, Hugo

doi:10.1002/mp.12215

Cited by 112 publications

(145 citation statements)

References 29 publications

(67 reference statements)

Supporting

Mentioning

141

Contrasting

Order By: Relevance

“…This arises because the sampled range errors in voxels of equal length are influenced by SPR errors and this aspect is discussed in section 4 of Ref. . This result also supports with the findings of Bornefalk who suggested that the intrinsic dimensionality of low‐Z elements at diagnostic energies is four.…”

Section: Discussionsupporting

confidence: 84%

“…That is,

Z_{normalmed} \equiv {\overset{σ}{true}}_{e, Z}^{- 1} false(σ_{normale, normalmed} false) .

For a postcalibration measurement, the estimator of

Z_{med}

is given by:

{\overset{Z}{true}}_{normalmed} = {false\sum}_{j = 1}^{J} c_{j} {normalΓ}^{j - 1}

where Γ is a quantity independent of electron density, being either the dual‐energy index or dual‐energy ratio, and

c_{j}

are calibration coefficients. In this study, we use J =3 to insure optimal robustness against noise . Based on this technique, the estimator of the ED is:

{\overset{ρ}{true}}_{normale} = \frac{u}{\sum_{l = 1}^{L} b_{l} Z_{normalmed}^{l - 1}},

where

b_{l}

are calibration coefficients.…”

Section: Methodsmentioning

confidence: 99%

“…The accuracy of the methods in SPR extraction is evaluated by calculating range uncertainties using the technique proposed by Bär et al., which combines Monte Carlo statistical sampling and WEPL‐based range prediction. For each method compared in this study, a set of SPR absolute errors obtained with the simulated CT data is used to create a cumulative density function (CDF) of water‐equivalent‐path‐length (WEPL) errors.…”

Section: Methodsmentioning

confidence: 99%

“…A general weakness of DECT‐based SPR extraction is its sensitivity to CT noise. This fact has been underlined by many authors in recent publications . Indeed, DECT techniques deal with magnified levels of noise inherent to the constraint of limiting imaging dose to the one of a SECT scan.…”

Section: Introductionmentioning

confidence: 99%

“…By using two distinctive energy spectra during a same acquisition, DECT allows separating the measured data in two independent variables, such the electron density (ED) and the effective atomic number (EAN), or basis material weights and density. Several calibration methods were published in the past few years to estimate SPR from measured DECT data, all yielding different levels of accuracy and practicability . It was also recently shown theoretically that multi‐energy CT (MECT) could improve SPR estimation by using more than two spectra in a noise‐free environment …”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Bayesian approach to solve proton stopping powers from noisy multi‐energy CT data

2017

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Discussionsupporting

confidence: 84%

“…That is,

Z_{normalmed} \equiv {\overset{σ}{true}}_{e, Z}^{- 1} false(σ_{normale, normalmed} false) .

For a postcalibration measurement, the estimator of

Z_{med}

is given by:

{\overset{Z}{true}}_{normalmed} = {false\sum}_{j = 1}^{J} c_{j} {normalΓ}^{j - 1}

where Γ is a quantity independent of electron density, being either the dual‐energy index or dual‐energy ratio, and

c_{j}

are calibration coefficients. In this study, we use J =3 to insure optimal robustness against noise . Based on this technique, the estimator of the ED is: