Premium
SU‐E‐T‐113: An Unified Imaging and Robotic Couch Quality Assurance Procedure and Automation Software
Author(s) -
Schreibmann E,
Elder E
Publication year - 2015
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4924474
Subject(s) - isocenter , imaging phantom , quality assurance , computer science , computer vision , artificial intelligence , software , orientation (vector space) , measure (data warehouse) , medical imaging , contouring , medical physics , nuclear medicine , computer graphics (images) , medicine , mathematics , data mining , external quality assessment , geometry , pathology , programming language
Purpose: To introduce a quality assurance (QA) procedure for the robotic 6DoF couches integrates the couch and imaging components QA procedures into a simple workflow. Method: Current QA procedure for the imaging systems is to carefully align a phantom at isocenter and measure in images acquired using the on‐board imaging (OBI) system if the phantom appears aligned with the isocenter. For the robotic couches a complementary procedure offsets a phantom from to measure the couch's ability to reposition the phantom at isocenter. We propose a two‐in –one procedure that starts with a phantom displaced with random positions and orientation and asks the imaging system to align it back at isocenter. Accuracy is measured by comparing the repositioned phantom againts the treatment field. A software module was also implemented to analyze, quantify and report the results. Results: The procedure has been implemented clinically using a SRS calibration phantom (VisionRT, London, UK) and the automated software version that is able to detect errors of less than 0.1 mm or 0.1 degrees. The software detects the markers locations with a combination of contouring and intensity filters to compute the shifts/rotations needed to reposition the phantom to it's ideal location using a interactive closest point algorithm to match markers segmentations as extracted from imaging with a mesh representation of their expected location according to the phantom specifications. Once the shifts are send to the couch for repositioning, the procedure is repeated to measure the residual error in imaging to measure robotic couch repositioning accuracy and the phantom location against the graticule is verified. Conclusions: A simplified QA procedure and the associated software are presented to streamline the quality assurance of imaging and 6D couch components into a single, easy‐to‐use procedure.