The use of exit detector sinograms to detect anatomical variations for patients extending beyond the TomoTherapy field of view: a feasibility study

• Published in: Medical physics (Lancaster) Vol. 39; no. 10; p. 6407
• Main Authors: Wooten, H Omar, Goddu, S Murty, Rodriguez, Vivian, Cates, Jeremy, Grigsby, Perry, Low, Daniel A
• Format: Journal Article
• Language: English
• Published: United States 01.10.2012
• Subjects: Feasibility Studies; Humans; Image Processing, Computer-Assisted - methods; Neoplasms - diagnostic imaging; Neoplasms - pathology; Neoplasms - radiotherapy; Phantoms, Imaging; Radiotherapy Planning, Computer-Assisted; Radiotherapy, Image-Guided - methods; Radiotherapy, Intensity-Modulated; Tomography, X-Ray Computed - methods
• ISSN: 0094-2405
• DOI: 10.1118/1.4754583

This work describes an independent method to use the TomoTherapy Hi-ART megavoltage CT imaging system for daily monitoring of anatomical changes of cancer patients whose anatomy extends beyond the imaging field of view. The imaging detector response to changes in attenuating media was measured using water-equivalent plastic. Weight loss was simulated using an anthropomorphic phantom and determining the system's ability to detect the weight loss. Layers of tissue-equivalent bolus were added to an anthropomorphic pelvis phantom and CT simulations of the phantom were conducted, one in which the phantom and bolus were both within the TomoTherapy imaging field of view, and another in which the couch was raised so that the bolus was outside the field of view. Gynecological treatment plans were developed using the TomoTherapy treatment planning system, and successive fractions of the plan were then delivered to the phantom. Weight loss was simulated by removing a 0.5 cm layer of bolus following each fraction. The exit detector sinograms were obtained from each fraction, and ratios of sinograms were calculated relative to a reference sinogram for which all bolus was in place. Histograms of ratio sinograms were determined and used to correlate with simulated weight loss. Exit detector sinograms and ratio histograms were also retrospectively analyzed for five patients all of whose anatomies extended beyond the imaging field of view and all of whom experienced weight variations exceeding 10% during treatment. Exit detector signal is well correlated to changes in attenuator thickness as demonstrated in both slab and anthropomorphic phantom geometries. Measured and expected signal increases agreed to within less than 2% for simulated weight loss on the anthropomorphic phantom. Exit detector signals for pelvic patients with significant weight loss variations were consistent with phantom measurements. The analysis of the ratio sinograms for the phantom measurements and real patients indicated that exit detector sinograms can be used to detect relative changes in patient anatomy for each fraction as a means of in vivo quality assurance.