Learning-based occupational x-ray scatter estimation

Abstract

Objective. During x-ray-guided interventional procedures, the medical staff is exposed to scattered ionizing radiation caused by the patient. To increase the staff’s awareness of the invisible radiation and monitor dose online, computational scatter estimation methods are convenient. However, such methods are usually based on Monte Carlo (MC) simulations, which are inherently computationally expensive. Yet, in the interventional environment, immediate feedback to the personnel is desirable. Approach. In this work, we propose deep neural networks to mitigate the computational effort of MC simulations. Our learning-based models consider detailed models of the (outer) patient shape and (inner) anatomy, additional objects in the room, and the x-ray tube spectrum to cover imaging settings encountered in real interventional settings. We investigate two cases of scatter prediction. First, we employ network architectures to estimate the full three-dimensional (3D) scatter distribution. Second, we investigate the prediction of two-dimensional (2D) intensity projections that facilitate the intra-procedural visualization. Main results. Depending on the dimensionality of the estimated scatter distribution and the network architecture, the mean relative error of each network is in the range of 12% to 14% compared to MC simulations. However, 3D scatter distributions can be estimated within 60ms and 2D distributions within 15ms. Significance. Overall, our method is suitable to support the online assessment of scattered ionizing radiation in the interventional environment and can help to lower the occupational radiation risk.

X-ray scatter in projection radiography

“…The overall shape of acquired scattered radiation in maps is in line with literature in standing chest PA and LAT protocols ( 1 , 26 ) and in supine AP pelvis, abdomen and bedside chest protocols ( 12 , 13 , 27 ) . However, the metallic structures of patient table attenuate radiation non-uniformly at different distances, which is most likely the reason for the altered scatter maps measured at the lowest height in the supine setups ( Figures 6 and 7 ).…”

A practical and robust method for beam blocker-based cone beam CT scatter correction

“…Monte Carlo simulation methods are considered as the 'gold standard' of scatter estimation. However, they are mainly limited by the calculation efficiency (Jia et al 2012, Bootsma et al 2015, Maul et al 2022. Compared with Monte Carlo methods, analytical methods can estimate scatter signal using a more simplified scatter model with a higher efficiency, but accurately modeling the scatter is still a big challenge (Zhao et al 2015, 2016, Nomura et al 2020.…”

“…This type of method performs well only when a decent registration between the current CBCT and the prior CT can be established , Shi et al 2017, Cui et al 2019, Cui et al 2021. Recent researches have shown the great potential of deep learning scatter correction methods, but their performance is heavily relied on the quality and quantity of the training data (Hansen et al 2018, Kida et al 2018, Maier et al 2018, Jiang et al 2019, Roser et al 2021, Maul et al 2022. On the other hand, the beam blocker methods directly measure the scatter signal through customized hardware by physically decoupling the scatter signal from the primary transmission signal.…”

Technical note: Characteristics of the energy spread kernels of scattered radiation in an x‐ray room

“…1 Knowing the energy spectrum of ionizing radiation inside an x-ray room is essential to optimize the design of radiation protection measures. [2][3][4][5][6][7][8][9][10] These measures must be taken for the protection of the patient 11,12 and the professionals inside the room [13][14][15][16][17][18][19][20][21][22] and the professionals and the public outside it. 23 In this work, Monte Carlo simulations are carried out to characterize the scattered radiation inside the x-ray room.…”

“…Knowing the energy spectrum of ionizing radiation inside an x‐ray room is essential to optimize the design of radiation protection measures 2–10 . These measures must be taken for the protection of the patient 11,12 and the professionals inside the room 13–22 and the professionals and the public outside it 23 …”