Han X.,Elekta Ltd
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2013
Structure segmentation of patient CT images is an essential step for radiotherapy planning but very tedious if done manually. Atlas-based auto-segmentation (ABAS) methods have shown great promise for getting accurate segmentation results especially when multiple atlases are used. In this work, we aim to further improve the performance of ABAS by integrating it with learning-based segmentation techniques. In particular, the Random Forests (RF) supervised learning algorithm is applied to construct voxel-wise structure classifiers using both local and contextual image features. Training of the RF classifiers is specially tailored towards structure border regions where errors in ABAS segmentation typically occur. The trained classifiers are applied to re-estimate structure labels at "ambiguous" voxels where labels from different atlases do not fully agree. The classification result is combined with traditional label fusion to achieve improved accuracy. Experimental results on H&N images and ribcage segmentation show clear advantage of the proposed method, which offers consistent and significant improvements over the baseline method. © 2013 Springer International Publishing. Source
Francis L.,Elekta Ltd
European Oncology and Haematology | Year: 2012
Radiation therapy has become among the safest and most effective methods for treating cancer. Technology continues to play a critical role in managing and supporting advanced treatment techniques and ensuring safer practice. For clinicians, the ability to tailor the technology to meet their specific oncology workflows is a challenge. This article discusses how, by being able to automate clinical workflow in the electronic medical record (EMR), not only delivers greater department efficiency but can help support better outcomes through optimising patient care, before, during and after treatment. © TOUCH MEDICAL MEDIA 2012. Source
Brooks R.,Elekta Ltd
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2011
Ultrasound data collected using a mechanical scanning probe naturally fits on a non-rectilinear coordinate frame. We report on the direct use of this non-rectilinear coordinate frame in a registration algorithm in the context of tracking intrafraction prostate motion during radiotherapy. Registration on the non-rectilinear frame performs as well as registration after reconstruction, and avoids the time demands and numerical noise introduced in a reconstruction process. © 2011 Springer-Verlag. Source
Elekta Inc. | Date: 2015-10-27
An adaptive therapy delivery system can receive imaging information including a volumetric image comprising a target such as a tumor or one or more other structures, and can receive imaging information corresponding to one or more imaging slices comprising different portions of the target, such as imaging slices acquired at different times after acquisition of the volumetric image. The system can spatially register information from an earlier-acquired image with a portion of the target included in a later-acquired one of the imaging slices. The system can then determine an updated location of the target indicated by the spatially-registered information. Using the updated location, the system can generate an updated therapy protocol to control delivery of a therapy beam.
Elekta Ltd | Date: 2014-10-27
Embodiments of the present invention provide phantoms, and associated methods of calibration which are suitable for use in both medical resonance imaging and radiographic imaging systems. A phantom for calibration of a medical imaging system, comprises a first component having a first outer shape, a portion of which defines part of at least one pocket; and a second component coupled to the first component and having a second outer shape, a portion of which defines another part of the at least one pocket. At least one of the first and second components comprises a reservoir, the reservoir having a shape at least a portion of which locates a centre of the at least one pocket.