2022
DOI: 10.1016/j.ejmp.2022.05.013
|View full text |Cite
|
Sign up to set email alerts
|

Deep learning-based in vivo dose verification from proton-induced secondary-electron-bremsstrahlung images with various count level

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
12
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
8

Relationship

3
5

Authors

Journals

citations
Cited by 13 publications
(12 citation statements)
references
References 34 publications
0
12
0
Order By: Relevance
“…This was because the images of prompt x-rays did not have a Bragg peak at the end of the beams, and thus the error was slightly larger in estimating the ranges. However, by using deep learning approaches, the distribution will be changed to dose distributions [41,42], possibly improving the accuracy of range estimation as well as estimating the dose distribution from the prompt x-ray images. Improving the accuracy in range estimation from the measured prompt x-ray images using such as deep learning or other methods is our future work to proceed.…”
Section: Discussionmentioning
confidence: 99%
See 1 more Smart Citation
“…This was because the images of prompt x-rays did not have a Bragg peak at the end of the beams, and thus the error was slightly larger in estimating the ranges. However, by using deep learning approaches, the distribution will be changed to dose distributions [41,42], possibly improving the accuracy of range estimation as well as estimating the dose distribution from the prompt x-ray images. Improving the accuracy in range estimation from the measured prompt x-ray images using such as deep learning or other methods is our future work to proceed.…”
Section: Discussionmentioning
confidence: 99%
“…Moreover, since the measured luminescence images represent the dose distribution, these image pairs of luminescence and prompt x-rays will be used for training in the deep learning of neural networks aimed at predicting dose distributions from prompt x-ray images [41,42], even for clinically used therapy beams. These are also our future work to proceed.…”
Section: Discussionmentioning
confidence: 99%
“…The difference between measured distribution and dose distribution was large for the images measured with the 4-mm-diameter collimator due to the lower spatial resolution. However, as shown previously, the distributions could be corrected to the dose distribution by using a neural network employing deep learning approaches [40,41].…”
Section: Discussionmentioning
confidence: 99%
“…Prompt X-ray imaging during irradiation of proton beam with an X-ray camera was easier to perform compared with prompt gamma photons because the energy of the prompt X-rays detected by the imaging method was low (20-60 keV). The disadvantages of prompt X-ray images are the absence of a Bragg peak in the images due to the low energy of the prompt X-rays produced from the secondary electrons of protons at the Bragg peak and higher attenuation of the X-rays in the subjects [27][28][29][30][31][32][33][34][35][36][37]. However, high contrast imaging of the beams has been achieved using an X-ray camera that employs an adequate scintillator and a pinhole with shield [27][28][29][30][31][32][33][34][35][36][37]39].…”
Section: Jinst 18 P07046mentioning
confidence: 99%
“…Other methods involve imaging of the positrons produced in a subject using a positron emission tomography (PET) system, a high-energy gamma camera, or a Compton camera [13][14][15][16][17][18][19][20][21][22][23][24][25][26]. Imaging using prompt secondary electron bremsstrahlung X-rays (hereinafter, simply prompt X-rays) emitted from a subject during proton beam irradiation is also a promising method for beam-range and beam-shape estimation [27][28][29][30][31][32][33][34][35][36][37].…”
Section: Introductionmentioning
confidence: 99%