A deep learning framework for prostate localization in cone beam CT‐guided radiotherapy

Abstract

Purpose: To develop a deep learning-based model for prostate planning target volume (PTV) localization on cone beam computed tomography (CBCT) to improve the workflow of CBCT-guided patient setup. Methods: A two-step task-based residual network (T 2 RN) is proposed to automatically identify inherent landmarks in prostate PTV. The input to the T 2 RN is the pretreatment CBCT images of the patient, and the output is the deep learning-identified landmarks in the PTV. To ensure robust PTV localization, the T 2 RN model is trained by using over thousand sets of CT images with labeled landmarks, each of the CTs corresponds to a different scenario of patient position and/or anatomy distribution generated by synthetically changing the planning CT (pCT) image. The changes, including translation, rotation, and deformation, represent vast possible clinical situations of anatomy variations during a course of radiation therapy (RT). The trained patient-specific T 2 RN model is tested by using 240 CBCTs from six patients. The testing CBCTs consists of 120 original CBCTs and 120 synthetic CBCTs. The synthetic CBCTs are generated by applying rotation/translation transformations to each of the original CBCT. Results: The systematic/random setup errors between the model prediction and the reference are found to be <0.25/2.46 mm and 0.14/1.41°in translation and rotation dimensions, respectively. Pearson's correlation coefficient between model prediction and the reference is higher than 0.94 in translation and rotation dimensions. The Bland-Altman plots show good agreement between the two techniques. Conclusions: A novel T 2 RN deep learning technique is established to localize the prostate PTV for RT patient setup. Our results show that highly accurate marker-less prostate setup is achievable by leveraging the state-of-the-art deep learning strategy.

Automated contour propagation of the prostate from pCT to CBCT images via deep unsupervised learning

“…The accuracy of the registration, and thus contour propagation, is adversely affected by image contents in one image that does not have correspondence in the other image, such as the presence/absence of bowel gas bladder filling on the CT images. 9 Besides, various image artifacts may ruin the global correspondence between pCT and CBCT. For example, metal artifacts may appear quite differently in different CT scans depending on the scanning modality and reconstruction algorithm.…”

“…For adaptive radiotherapy, an indispensable step toward the realization of this type of adaptive replanning is the segmentation of CBCT images. The accuracy of the registration, and thus contour propagation, is adversely affected by image contents in one image that does not have correspondence in the other image, such as the presence/absence of bowel gas bladder filling on the CT images 9 . Besides, various image artifacts may ruin the global correspondence between pCT and CBCT.…”

A Deep Unsupervised Learning Model for Artifact Correction of Pelvis Cone-Beam CT

“…As compared with planning CT (PCT), the image quality of CBCT is relatively poor, such as low signal-to-noise ratio, motion, and fringe artifacts ( 11 , 12 ). The error of CT value between PCT and CBCT may be as large as 200 Hounsfield unit (HU) ( 13 , 14 ).…”

Breast Cancer Patient Auto-Setup Using Residual Neural Network for CT-Guided Therapy

“…Researches have demonstrated that increasing the delivered radiation dose to the breast cancer will improve the disease control, especially for the patients who suffer from the advanced disease. Image segmentation and localization technique will improve the quality of the treatment [1][2][3][4][5][6][7][8]. However, with the current tumor localization methods, increasing the delivered dose to the breast or the heart will result in a higher dose to the surrounding normal tissues.…”

“…Different from the registration approaches based on the image intensity, in this paper, we establish a patient-specific deep learning model to achieve the breast auto-setup. The setup errors between the planning CT (pCT) and the daily CT is obtained by aligning the corresponding breast landmarks which are outputted from the deep learning model [8,[11][12][13]. It can address the drawbacks in the traditional methods through the following contributions: 1) Instead of requiring a large patient dataset, the training data in the article is generated using a patient-specific data augmentation strategy, which can account for the changes of the image content between the daily CT and the pCT.…”