A feasibility study for in vivo treatment verification of IMRT using Monte Carlo dose calculation and deep learning-based modelling of EPID detector response

Abstract

Background This paper describes the development of a predicted electronic portal imaging device (EPID) transmission image (TI) using Monte Carlo (MC) and deep learning (DL). The measured and predicted TI were compared for two-dimensional in vivo radiotherapy treatment verification. Methods The plan CT was pre-processed and combined with solid water and then imported into PRIMO. The MC method was used to calculate the dose distribution of the combined CT. The U-net neural network-based deep learning model was trained to predict EPID TI based on the dose distribution of solid water calculated by PRIMO. The predicted TI was compared with the measured TI for two-dimensional in vivo treatment verification. Results The EPID TI of 1500 IMRT fields were acquired, among which 1200, 150, and 150 fields were used as the training set, the validation set, and the test set, respectively. A comparison of the predicted and measured TI was carried out using global gamma analyses of 3%/3 mm and 2%/2 mm (5% threshold) to validate the model's accuracy. The gamma pass rates were greater than 96.7% and 92.3%, and the mean gamma values were 0.21 and 0.32, respectively. Conclusions Our method facilitates the modelling process more easily and increases the calculation accuracy when using the MC algorithm to simulate the EPID response, and has potential to be used for in vivo treatment verification in the clinic.

AAPM Task Group Report 307: Use of EPIDs for Patient‐Specific IMRT and VMAT QA

“…A work by Zhang et al. describes the development of a predicted EPID transmission image using Monte Carlo and deep learning algorithms for potential use for in vivo treatment verification 145 . Nyflot et al.…”

“…Valdes et al developed a machine learning algorithm to correlate the characteristics of IMRT plan and delivery characteristic and the corresponding gamma passing rate for a variety of QA devices 144. A work by Zhang et al describes the development of a predicted EPID transmission image using Monte Carlo and deep learning algorithms for potential use for in vivo treatment verification 145. Nyflot et al investigated a deep learning approach to classify the presence or absence of intentionally introduced treatment delivery errors from IMRT PSQA.…”

Artificial intelligence assists precision medicine in cancer treatment

“…The DL model based on one of the subsets of AI can assist radiotherapy or oncology doctors in accurately outlining tumor targets, reducing the time doctors take to manually segment images as well as reducing the variation between observers (159)(160)(161)(162)(163). The model can predict and verify the therapeutic dose, and allows for the dose prescription to be changed in time to reduce the impact on the surrounding normal tissue, prevent unnecessary radiation, and reduce the occurrence of adverse reactions (164,165). The model can evaluate the efficacy of radiotherapy and chemotherapy, as well as the therapeutic response, so as to achieve better-personalized prescriptions for patients (166)(167)(168).…”

A Novel Approach for Position Verification and Dose Calculation through Local MVCT Reconstruction

“…Our team conducted earlier research on pre-treatment dose verification based on EPID [ 25 ] and in vivo dose verification methods [ 25 , 26 ], discovering that EPID transmission images contain patient anatomical information, which can be utilized for reconstructing the three-dimensional structure of patients. Furthermore, the radiation source used for EPID image acquisition is the same as that used during patient treatment, which addresses the limitations of CBCT for positioning verification to some extent.…”