Xun Jia

Predicting treatment plan approval probability for high-dose-rate brachytherapy of cervical cancer using adversarial deep learning

Abstract Objective. Predicting the probability of having the plan approved by the physician is important for automatic treatment planning. Driven by the mathematical foundation of deep learning that can use a deep neural network to represent functions accurately and flexibly, we developed a deep-learning framework that learns the probability of plan approval for cervical cancer high-dose-rate brachytherapy (HDRBT). Approach. The system consisted of a dose prediction network (DPN) and a plan-approval probability network (PPN). DPN predicts organs at risk (OAR) D 2cc and CTV D 90% of the current fraction from the patient’s current anatomy and prescription dose of HDRBT. PPN outputs the probability of a given plan being acceptable to the physician based on the patients anatomy and the total dose combining HDRBT and external beam radiotherapy sessions. Training of the networks was achieved by first training them separately for a good initialization, and then jointly via an adversarial process. We collected approved treatment plans of 248 treatment fractions from 63 patients. Among them, 216 plans from 54 patients were employed in a four-fold cross validation study, and the remaining 32 plans from other 9 patients were saved for independent testing. Main results. DPN predicted equivalent dose of 2 Gy for bladder, rectum, sigmoid D 2cc and CTV D 90% with a relative error of 11.51% ± 6.92%, 8.23% ± 5.75%, 7.12% ± 6.00%, and 10.16% ± 10.42%, respectively. In a task that differentiates clinically approved plans and disapproved plans generated by perturbing doses in ground truth approved plans by 20%, PPN achieved accuracy, sensitivity, specificity, and area under the curve 0.70, 0.74, 0.65, and 0.74. Significance. We demonstrated the feasibility of developing a novel deep-learning framework that predicts a probability of plan approval for HDRBT of cervical cancer, which is an essential component in automatic treatment planning.

Definitive radiation for advanced cervix cancer is not associated with vaginal shortening—a prospective vaginal length and dose correlation

Prospectively measure change in vaginal length after definitive chemoradiation (C-EBRT) with Intracavitary Brachytherapy (ICBT) for locally advanced cervix cancer (LACC) and correlate with vaginal dose (VD). Twenty one female patients with LACC receiving C-EBRT and ICBT underwent serial vaginal length (VL) measurements. An initial measurement was made at the time of the first ICBT procedure and subsequently at 3 month intervals up to 1 year post radiation. The vagina was contoured as a 3-dimensional structure for each brachytherapy plan. The difference in VL before and at least 6 months after the last fraction of brachytherapy was considered as an indicator of toxicity. The mean initial VL was 8.7 cm (6.5-12) with median value of 8.5 cm. The mean VL after 6 months was 8.6 cm (6.5-12) and VL change was not found to be statistically significant. The median values (interquartile ranges) for vaginal D0.1cc, D1cc, and D2cc were 129.2 Gy (99.6-252.2), 96.9 Gy (84.2-114.9), and 89.6 Gy (82.4-102.2), respectively. No significant correlation was found between vaginal length change and the dosimetric parameters calculated for all patients. Definitive C-EBRT and ICBT did not significantly impact VL in this prospective cohort probably related to acceptable doses per ICRU constraints. Estimate of vaginal stenosis and sexual function was not performed in this cohort which is a limitation of this study and which we hope to study prospectively going forward.