Search

Manual image segmentation is a time-consuming task routinely performed in radiotherapy to identify each patient's targets and anatomical structures. The efficacy and safety of the radiotherapy plan requires accurate segmentations as these regions of interest are generally used to optimize and assess the quality of the plan. However, reports have shown that this process can be subject to significant inter- and intraobserver variability. Furthermore, the quality of the radiotherapy treatment, and subsequent analyses (ie, radiomics, dosimetric), can be subject to the accuracy of these manual segmentations. Automatic segmentation (or auto-segmentation) of targets and normal tissues is, therefore, preferable as it would address these challenges. Previously, auto-segmentation techniques have been clustered into 3 generations of algorithms, with multiatlas based and hybrid techniques (third generation) being considered the state-of-the-art. More recently, however, the field of medical image segmentation has seen accelerated growth driven by advances in computer vision, particularly through the application of deep learning algorithms, suggesting we have entered the fourth generation of auto-segmentation algorithm development. In this paper, the authors review traditional (nondeep learning) algorithms particularly relevant for applications in radiotherapy. Concepts from deep learning are introduced focusing on convolutional neural networks and fully-convolutional networks which are generally used for segmentation tasks. Furthermore, the authors provide a summary of deep learning auto-segmentation radiotherapy applications reported in the literature. Lastly, considerations for clinical deployment (commissioning and QA) of auto-segmentation software are provided., (Copyright © 2019 Elsevier Inc. All rights reserved.)

Radiation therapy is an effective, personalized cancer treatment that has benefited from technological advances associated with the growing ability to identify and target tumors with accuracy and precision. Given that these advances have played a central role in the success of radiation therapy as a major component of comprehensive cancer care, the American Society for Radiation Oncology (ASTRO), the American Association of Physicists in Medicine (AAPM), and the National Cancer Institute (NCI) sponsored a workshop entitled "Technology for Innovation in Radiation Oncology," which took place at the National Institutes of Health (NIH) in Bethesda, Maryland, on June 13 and 14, 2013. The purpose of this workshop was to discuss emerging technology for the field and to recognize areas for greater research investment. Expert clinicians and scientists discussed innovative technology in radiation oncology, in particular as to how these technologies are being developed and translated to clinical practice in the face of current and future challenges and opportunities. Technologies encompassed topics in functional imaging, treatment devices, nanotechnology, and information technology. The technical, quality, and safety performance of these technologies were also considered. A major theme of the workshop was the growing importance of innovation in the domain of process automation and oncology informatics. The technologically advanced nature of radiation therapy treatments predisposes radiation oncology research teams to take on informatics research initiatives. In addition, the discussion on technology development was balanced with a parallel conversation regarding the need for evidence of efficacy and effectiveness. The linkage between the need for evidence and the efforts in informatics research was clearly identified as synergistic., (Copyright © 2015 Elsevier Inc. All rights reserved.)