Synthetic CT-aided Online CBCT Multi-Organ Segmentation for CBCT-guided Adaptive Radiotherapy of Pancreatic Cancer

Abstract

Median copy number per mL was 41.7 in the total cohort, 1054.2 (range 17.5-161083.3) in OPX, 479.2 (range 22.9-38500.0) in AC, and 39.2 (range 12.9-39666.7) in UC patients. There was no association between ctHPVDNA detectability and age at diagnosis, sex, smoking history, T stage, N stage, or M stage. Among patients with cervical cancer, ctHPVDNA was detectable in 80% of patients with squamous or adenosquamous histology vs 25% of patients with adenocarcinoma (p Z 0.0314). The median follow-up for all patients was 27 months (IQR 17.9-34.3 months). Recurrence at any site occurred in 14 of 71 patients (19.7%). Compared to patients with undetectable ctHPVDNA, detectable ctHPVDNA was not associated with disease recurrence (hazard ratio 0.925, 95% CI 0.266-3.217, p Z 0.902). There was no association between ctHPVDNA copy number and recurrence when stratifying by <200 copies per mL versus >200 copies per mL. Conclusion: We demonstrate the feasibility of pre-treatment ctHPVDNA detectability in patients with squamous or adenosquamous carcinoma of the oropharynx, anal canal, and uterine cervix. This is the first validation of a ctHPVDNA assay in anal canal and cervical cancer. Further investigation of ctHPVDNA levels pre-treatment, post-treatment, and during surveillance in these malignancies is warranted.

Deep learning methods for enhancing cone‐beam CT image quality toward adaptive radiation therapy: A systematic review

“…demonstrated Dice score agreement greater than 0.89 for the pelvic region 46,47 . Recently, segmentation‐specific studies have compared the performance of CBCT‐derived sCT images for abdominal segmentations, reporting Dice scores above 0.8 125,126 . The validation of sCT‐based auto‐segmentation for all anatomic regions is an important step toward online ART.…”

Synthetic CT‐aided multiorgan segmentation for CBCT‐guided adaptive pancreatic radiotherapy

“…A leave‐one‐out cross‐validation approach was applied wherein, for each of 35 experiments, images belonging to each of the 35 patients were sequentially omitted during training for use as test data. The cycleGAN was trained and tested as detailed in our previous studies 46,58 . Here, the learning rate for Adam optimizer was set to 2 × 10 −4 .…”

“…The cycle-GAN was trained and tested as detailed in our previous studies. 46,58 Here, the learning rate for Adam optimizer was set to 2 × 10 −4 . The training was stopped after 150 000 iterations.…”

“…43,44 Synthetic image-aided segmentation has been demonstrated as a promising method for improving the accuracy of segmentation in our previous studies. [45][46][47] Compared to these previous investigations, in this work, we use the strategy of image synthesis to improve the image quality of the original CBCT. We hypothesize that with the improved image quality of sCT compared to the original CBCT, the segmentation performance can be improved.…”

“…In the second step, a mask scoring regional convolutional neural network (MS R-CNN) is applied to the sCT to obtain final contours of multiple organs simultaneously. Compared to our previous studies, [45][46][47] MS R-CNN first detects the organs' regions of interest (ROIs) and then segments organs' binary mask within these ROIs.One hundred twenty sets of CBCT images from patients with pancreatic cancer were used to assess the performance of our proposed method.…”