Abstract
An accurate detection of the radiation field is crucial to 3D conformal radiotherapy (3D‐CRT). Automated techniques to detect the field edges on double exposure portal images have previously focused on thresholding techniques. In this paper, we present a new approach based on a curve propagation technique (the Fast Marching method) which proves to be more effective at detecting the radiation field than its thresholding counterpart. The comparison of both techniques in terms of computational speed and effectiveness of the detection is presented using complex images with non‐homogeneous intensity levels inside the radiation field, and gradual variations in intensity level at the field boundaries. Results show that our Fast Marching method is easier to automate, and converges faster to the boundaries of the segmented radiation field. The computation time of the Fast Marching technique is five times faster in typical portal images.PACS numbers: 87.53.Oq, 87.57.Nk, 87.57.‐s.
Highlights
In three-dimensional conformal radiotherapy (3D-CRT), multiple high-energy radiation beams are delivered to tumors to kill cancer cells
The portal verification of the positioning of these radiation fields is carried out by comparing the location of the treated portal image taken during delivery with the reference simulation, which is an image taken on a simulator or a digitally reconstructed radiograph (DRR).(1-5) Developing an efficient method for an automated verification of the treatment portal localization is crucial to the quality assurance of conformal radiotherapy.[6,7] If localization accuracy is improved, it is possible to envision applying higher dosage to the tumor with exclusion of excess to normal tissues
We present an enhanced version of the gradient which includes ‘a priori’ information to remove additional noise, which suggests that the Fast Marching method has the potential to be successful in detecting radiation field on a large variety of images, from simple to complex
Summary
In three-dimensional conformal radiotherapy (3D-CRT), multiple high-energy radiation beams are delivered to tumors to kill cancer cells. Thresholding techniques associated with gradient information provide adequate segmentation results in images with homogeneous density This paper presents a modification of the automatic detection of the threshold proposed by Wang et al[8] to significantly reduce the computation time They fail to find the correct edges of the radiation field when the intensity variation is weaker on the boundaries of the radiation field than inside the treated area. We present an enhanced version of the gradient which includes ‘a priori’ information to remove additional noise, which suggests that the Fast Marching method has the potential to be successful in detecting radiation field on a large variety of images, from simple to complex. We conclude that the Fast Marching technique is more efficient both computationally and qualitatively, and has more automation potential
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
