Applied Radiation and Isotopes

An automatic patient-specific quality assurance with a novel DVH scoring algorithm for volumetric modulated arc therapy of cervical cancer

To develop a novel DVH scoring algorithm to predict and classify patient-specific quality assurance (PSQA) results by different DVH metrics automatically and efficiently. A total of 200 cervical cancer patients who treated by Infinity (109 cases) and Synergy (91 cases) linear accelerators underwent volumetric modulated arc therapy (VMAT) from 2019 to 2022 were enrolled and used as technical testing (TT) and technical validation (TV) datasets, respectively, which was then randomly divided into training, validation and testing set at a ratio of 7:1:2. PSQA dose distributions were predicted using U-shape-like network with skip-connection modules (called T-Net) with the input of CT and plan dose distributions. A novel weight-based DVH scoring (WDS) algorithm was developed and trained to classify "pass" or "fail" (PoF) of PSQA results based on the dose errors (DEs) and volumetric errors (VEs) calculated between predicted and planned DVHs. T-Net achieved a best performance in predicting PSQA dose distributions in comparison with other deep learning models. The WDS method achieved a sensitivity, specificity and accuracy of 100.00 %, 50.00 %, 0.955, and 100.00 %,33.33 %, 0.890 in TT and TV, respectively, which was better than models of random forest (RF) and support vector machines (SVM) with an accuracy of 0.909, 0.833 and 0.864, 0.722 in TT and TV, respectively. The threshold DVH score for 22 and 18 validation patients were 49.62 and 57.62 in the TT and TV with a precision, recall rate and F1 score of 0.952, 1, 0.976 and 0.882, 1, 0.938, respectively. The suggested novel WDS algorithm can improve the accuracy and efficiency of classifying the PoF of PSQA objectively and automatically.

Preclinical studies and absorbed dose estimation of [89Zr]Zr-DFO-Bevacizumab for PET imaging of VEGF-expressing tumors

This study aimed to carry out the preclinical studies of [

Quantified difference of the collapsed cone convolution (CCC) and Monte Carlo (MC) algorithms based on DVH and gamma analysis for cervical cancer radiation therapy

To quantify the difference between the (collapsed cone convolution) CCC algorithm and the (Monte Carlo) MC algorithm and remind that the planners should pay attention to some possible uncertainties of the two algorithms when employing the two algorithms. Thirty patients' cervical cancer VMAT plans were designed with a Pinnacle TPS (Philips) and divided equally into two groups: the simple group (SG, target volume was only the PTV) and the complex group (CG, target volume included the PTV and PGTV). The plans from the Pinnacle TPS were transferred to the Monaco TPS (Elekta). The plans' parameters all remained unchanged, and the dose was recalculated. Gamma passing rates (GPRs) obtained from dose distribution from Pinnacle TPS compared with that from Monaco TPS with SNC software based on three triaxial planes (transverse, sagittal and coronal). GPRs and DVH were used to quantify the difference between the CCC algorithm in pinnacle TPS and the MC algorithm in Monaco TPS. Among the statistical dose indexes in DVHs from the Pinnacle and Monaco TPSs, there were 7(7/15) dose indexes difference with statistically significant differences in the SG, and 10(10/18) dose indexes difference with statistically significant differences in the CG. With 3%/3 mm criterion, the most (5/6) GPRs were greater than 95% from the SG and CG. But with 2%/2 mm criterion, the most (5/6) GPRs were less than 90% from the two groups. In addition, we found that GPRs were also related to the selected triaxial planes and the complexity of the plan (GPRs varied with the SG and CG). Obvious difference between the CCC and MC algorithms from Pinnacle and Monaco TPS. DVH maybe better than 2D gamma analysis on quantifying difference of the CCC and MC algorithms. Some attention should be paid to the uncertainty of the TPS algorithm, especially when the indicator on the DVH is at the critical point of the threshold value, because the algorithm used may overestimate or underestimate the DVH indicator.

Evaluating dose distributions of high dose rate 60Co brachytherapy in an asymmetric tumor: A comparison of different designs of vaginal cylindrical applicators

The aim of this study is to evaluate organ-at-risk (OAR) doses obtained from different designs of intracavitary cylinder applicators during high-dose-rate (HDR) brachytherapy of an asymmetric vaginal tumor. Dose distributions around five cylinder applicators were obtained using the Geant4 toolkit. Three of these applicators are commercially available: single-channel, multi-channel, and shielded. Additionally, two proposed sub-configurations were evaluated: (I) a combination of the multichannel applicator with a central shield and (II) a dynamically modulated cylinder (DMC). The dose distributions of the proposed applicators were compared to those of the single-channel cylinder applicator. The results showed that using a cylindrical applicator with a higher degree of freedom in plan optimization can improve OAR, sparing up to 60% for the rectum and 20% for the bladder. In conclusion, this work suggests using a new design of a cylindrical applicator to enhance the therapeutic ratio of brachytherapy for asymmetric and irregularly shaped vaginal tumors.

Comparison between MCNP and planning system in brachytherapy of cervical cancer

Absorbed doses in uterus during brachytherapy were calculated with MCNP in relevant points and compared with planning system for one patients. MCNP was applied with two different humanoid phantoms in input, ORNL and voxel models, which represent human body in mathematical way. Good agreement between both phantoms, as well as, between MCNP and planning system were found. In addition the doses in critical organs (bladder and colon in this kind of therapy), were calculated and compared with maximal doses in these organs obtained from planning system for 15 other patients. MCNP doses agree well with planning system in points of uterus for those 15 patients, where radioactive source is used to apply. However, there are systematical discrepancies between doses in colon and bladder obtained by MCNP and planning system.

A phamtom study: In vivo rectal dosimetry of high dose rate brachytherapy in cervical cancer

The purpose of this study was to perform in-vivo dosimetry using a diode rectal dosimeter in phantom and compare the dose delivered to the rectum between the dose measured by the diode dosimeter and the dose calculated by the treatment planning system in cervical cancer. The PTW T9112 diode detector calibrations were performed to find the correction factor. Then the calibrated diode detector was used to measure the radiation dose received in the rectum area in the in-house pelvic phantom. An Iridium-192 source was loaded into the phantom with 7 Gy, the measurements were 3 times per treatment plan, with 15 total plans studied. The average cumulative charge (nC) of each plan was converted to the absorbed dose (mGy) for comparison with the treatment planning system. Finally, to test the hypothesis that an absorbed dose from the detector and the treatment planning system were not significantly different, dependent t-test statistical analysis was applied with p-value <0.05. For distance and direction correction factors, we found that the factors were approximately 1 at 5 cm and 180°. The percentage differences of radiation dose between the diode dosimeter and the treatment planning system were between -3.3 and 4.1%. Statistical analysis revealed that the doses from the detector and the treatment planning system were not statistically significant different. The comparison showed that the percent difference between diode dosimeter and treatment planning system was acceptable to perform the in vivo dosimetry in brachytherapy. Therefore, the diode detector may be a suitable candidate for a treatment verification system in cervical cancer brachytherapy to prevent the dose delivery errors that directly affect the prognosis and may cause complications for the patient.

Study of the application of the deformed dose summation (DDS) method in brachytherapy for advanced cervical cancer and analysis of regional dosing in recurrent cervical cancer

This study evaluated the cumulative brachytherapy (BT) dose in cervical cancer using different methods and explored using deformed dose summation (DDS) techniques for assessing doses to targets and organs at risk (OAR). A total of 41 patients with cervical cancer who underwent BT were retrospectively analyzed, the cumulative dose, measured as the equivalent dose in 2 Gy fractions, was calculated using dose-volume histogram (DVH) superposition and DDS methods. The dice similarity coefficient (DSC), Jaccard coefficient (JC), and mean distance to agreement (MDA) were utilized to evaluate the deformable image registration (DIR) accuracy. Furthermore, the difference between the target and OAR doses obtained through the two methods was calculated and compared. The Spearman method was employed to analyze dose differences and geometric correlations, comparing the dose in the relapsing area with that in the post-fusion high-risk clinical target volume (HR-CTV) in patients experiencing relapse to identify potential associations. In evaluating deformable registration, the registration outcomes for both the bladder and rectum were deemed satisfactory. The DSC, JC, and MDA values for the bladder and rectum were calculated as follows: 0.89 ± 0.07, 0.81 ± 0.09, 2.14 ± 1.44 mm and 0.80 ± 0.08, 0.67 ± 0.10, 2.28 ± 1.80 mm, respectively. Compared to DVH superposition, DDS techniques reduced the average cumulative doses of the D2cc and D0.1 cc of bladder by 11.4 % and 8.2 %, respectively; and the doses of the D2cc and D0.1 cc of rectum to be decreased by 9.7 % and 14.1 %, respectively. Additionally, the D90 of HR-CTV and Intermediate risk CTV (IR-CTV) were reduced by 6.60 ± 3.70 Gy and 4.55 ± 2.70 Gy, respectively. No significant correlation was observed between DSC, JC, and MDA and the dose difference. No correlation was observed between the relapsing regional dose and the dose parameters related to the HRCTV following fusion. Calculating the cumulative dose using DVH parameters is a conservative approach that may limit target dose enhancement. During planning, DIR can guide clinicians in selecting the target dose and visually display the cumulative dose distribution in the cervical target area.

Deep learning-based automatic dose optimization for brachytherapy

The purpose of this study is to determine the best dose processing method for deep learning-based dose prediction in brachytherapy (BT), as well as to investigate the feasibility of using the inverse dose optimization algorithm to improve treatment planning quality. BT data from 186 patients with cervical cancer were retrospectively collected. The data were divided into three sets: training, validation, and test, with a ratio of 150:18:18. The dose data was normalized using square-root transformation normalization, logarithmic normalization, and linear normalization. For dose distribution prediction, a 3D U-Net architecture was used. The predicted results were compared to unprocessed dose data. The four groups of dose predictions were assessed using the Dice similarity coefficient (DSC), conformity index (CI), and homogeneity index (HI). The group with the best overall performance was chosen, and the dose prediction results were fed into a gradient-based planning optimization (GBPO) algorithm for additional optimization. The target D90 % was normalized to 6 Gy. The D1cc and D2cc of the OARs were compared prior to and following optimization. The dose prediction method using unprocessed doses produced the best overall performance on the DSC, CI, and HI metrics. The (DSC, CI, HI) values for unprocessed dose, square-root transformation normalized, log normalized, and linear normalized were (0.94, 0.74, 0.49), (0.93, 0.72, 0.50), (0.91, 0.71, 0.45) and (0.90, 0.71, 0.47), respectively. The predicted dose results for the unprocessed dose group were further optimized by the GBPO algorithm. The outcomes demonstrated that the (D1cc, D2cc) values for the bladder, rectum, and sigmoid decreased by (2.11 %, 2.09 %), (2.62 %, 2.14 %) and (3.16 %, 2.98 %), respectively, and were statistically significant (p  0.05). When using deep learning for BT dose prediction in the 3D U-Net model with the cervical cancer BT data used in this study, dose normalization processing is not recommended The predicted dose can be further optimized using inverse dose optimization algorithms to improve the treatment plan's quality.