Comparison of computed tomography- and magnetic resonance imaging-based target delineation for cervical cancer brachytherapy

Wang, Fang; Bu, Luyi; Wu, Qun; Jiang, Xiyi; Wu, Lingyun; Yu, Li; Xi, Bin; Lu, Zhongjie; Yan, Senxiang

doi:10.5114/jcb.2020.98117

“…More consistent and well-curated training data using MRI will eventually lead to a better segmentation performance of DL-based CTV segmentation. Wang et al 39 demonstrated that CT scan cannot clearly distinguish between structures of residual GTV, cervix, uterus, and vagina. Also, CT provides a poor definition of parametrial tumor infiltration.…”

Section: Discussionmentioning

confidence: 99%

Fully automated segmentation of clinical target volume in cervical cancer from magnetic resonance imaging with convolutional neural network

Zabihollahy

¹

,

Viswanathan

²

,

Schmidt

³

et al. 2022

J Applied Clin Med Phys

View full text Add to dashboard Cite

Contouring clinical target volume (CTV) from medical images is an essential step for radiotherapy (RT) planning. Magnetic resonance imaging (MRI) is used as a standard imaging modality for CTV segmentation in cervical cancer due to its superior soft-tissue contrast. However, the delineation of CTV is challenging as CTV contains microscopic extensions that are not clearly visible even in MR images, resulting in significant contour variability among radiation oncologists depending on their knowledge and experience. In this study, we propose a fully automated deep learning-based method to segment CTV from MR images. Methods: Our method begins with the bladder segmentation, from which the CTV position is estimated in the axial view. The superior-inferior CTV span is then detected using an Attention U-Net. A CTV-specific region of interest (ROI) is determined, and three-dimensional (3-D) blocks are extracted from the ROI volume. Finally, a CTV segmentation map is computed using a 3-D U-Net from the extracted 3-D blocks. Results: We developed and evaluated our method using 213 MRI scans obtained from 125 patients (183 for training, 30 for test). Our method achieved (mean ± SD) Dice similarity coefficient of 0.85 ± 0.03 and the 95th percentile Hausdorff distance of 3.70 ± 0.35 mm on test cases, outperforming other stateof -the-art methods significantly (p-value < 0.05). Our method also produces an uncertainty map along with the CTV segmentation by employing the Monte Carlo dropout technique to draw physician's attention to the regions with high uncertainty, where careful review and manual correction may be needed. Conclusions: Experimental results show that the developed method is accurate, fast, and reproducible for contouring CTV from MRI, demonstrating its potential to assist radiation oncologists in alleviating the burden of tedious contouring for RT planning in cervical cancer.

show abstract

“…The American Brachytherapy Society (ABS) conducted a survey of image guided cervical brachytherapy practices in the United States, 95% of participants always use CT, and 34% always use MRI for dose specification to target (79). However, CT-based target contouring is not recommended because in CT images the target is overestimated in size in comparison to the target from MRI images (80).…”

Section: D-based Brachytherapymentioning

confidence: 99%

Improving Reproducibility and Efficiency of Interstitial HDR Brachytherapy Through MRI-based Targeted Innovations

Shaaer

2023

Preprint

0

View full text Add to dashboard Cite

Interstitial high dose rate (HDR) brachytherapy is a well-established treatment for prostate and gynecological cancers. HDR brachytherapy involves delivery of a high dose of radiation to the target with minimum dose to the surrounding tissue. Magnetic resonance imaging (MRI) has been explored as a main imaging modality for accurate target identification. Superior soft tissue resolution has been reported with MRI, increasing the accuracy of brachytherapy procedure. Clinical HDR brachytherapy workflows employing MRI are seeing increases in adoption. However, these require additional resources and patient transfers, which increases the planning time and the risk of catheters displacement. In addition, interstitial catheter reconstruction during interstitial brachytherapy procedure is challenging. Thus, accurate and efficient methods of incorporating the MRI imaging in interstitial HDR brachytherapy procedures need to be considered. In this thesis, a robust MRI-based algorithm and workflow were developed for treatment planning of interstitial HDR brachytherapy for prostate and gynecological cancers. An in-house MRI-to Ultrasound image registration algorithm was developed and evaluated for accurate identification of dominant intraprostatic lesions. The dosimetric impact of the registration was also assessed. Furthermore, an in-house MR-line marker was used in evaluating the feasibility of MR-only interstitial gynecological HDR brachytherapy workflow. The impact of the MR-only workflow on the target coverage and normal tissue sparing was evaluated. The dosimetric difference between the conventional approach and the proposed MR-only approaches was also compared and reported. Finally, a deep learning-assisted algorithm was developed for robust catheter reconstruction. The algorithm was benchmarked against the manual reconstruction and their efficiency was evaluated. Based on the results, the MRI-based innovations developed in this work are feasible for implementation in clinical brachytherapy workflows for treatment planning of prostate and gynecological cancers.

show abstract

“…The American Brachytherapy Society (ABS) conducted a survey of image guided cervical brachytherapy practices in the United States, 95% of participants always use CT, and 34% always use MRI for dose specification to target (79). However, CT-based target contouring is not recommended because in CT images the target is overestimated in size in comparison to the target from MRI images (80).…”

Section: D-based Brachytherapymentioning

confidence: 99%

Improving Reproducibility and Efficiency of Interstitial HDR Brachytherapy Through MRI-based Targeted Innovations

Shaaer

2023

Preprint

0

View full text Add to dashboard Cite

Interstitial high dose rate (HDR) brachytherapy is a well-established treatment for prostate and gynecological cancers. HDR brachytherapy involves delivery of a high dose of radiation to the target with minimum dose to the surrounding tissue. Magnetic resonance imaging (MRI) has been explored as a main imaging modality for accurate target identification. Superior soft tissue resolution has been reported with MRI, increasing the accuracy of brachytherapy procedure. Clinical HDR brachytherapy workflows employing MRI are seeing increases in adoption. However, these require additional resources and patient transfers, which increases the planning time and the risk of catheters displacement. In addition, interstitial catheter reconstruction during interstitial brachytherapy procedure is challenging. Thus, accurate and efficient methods of incorporating the MRI imaging in interstitial HDR brachytherapy procedures need to be considered. In this thesis, a robust MRI-based algorithm and workflow were developed for treatment planning of interstitial HDR brachytherapy for prostate and gynecological cancers. An in-house MRI-to Ultrasound image registration algorithm was developed and evaluated for accurate identification of dominant intraprostatic lesions. The dosimetric impact of the registration was also assessed. Furthermore, an in-house MR-line marker was used in evaluating the feasibility of MR-only interstitial gynecological HDR brachytherapy workflow. The impact of the MR-only workflow on the target coverage and normal tissue sparing was evaluated. The dosimetric difference between the conventional approach and the proposed MR-only approaches was also compared and reported. Finally, a deep learning-assisted algorithm was developed for robust catheter reconstruction. The algorithm was benchmarked against the manual reconstruction and their efficiency was evaluated. Based on the results, the MRI-based innovations developed in this work are feasible for implementation in clinical brachytherapy workflows for treatment planning of prostate and gynecological cancers.

show abstract

Comparison of computed tomography- and magnetic resonance imaging-based target delineation for cervical cancer brachytherapy

Cited by 3 publications

References 35 publications

Fully automated segmentation of clinical target volume in cervical cancer from magnetic resonance imaging with convolutional neural network

Fully automated segmentation of clinical target volume in cervical cancer from magnetic resonance imaging with convolutional neural network

Improving Reproducibility and Efficiency of Interstitial HDR Brachytherapy Through MRI-based Targeted Innovations

Improving Reproducibility and Efficiency of Interstitial HDR Brachytherapy Through MRI-based Targeted Innovations

Contact Info

Product

Resources

About