Frédéric Gaspoz1, Pascal Monnin1, David Petter2, Jean Plé2, Sandrine Ding3. 1. Department of Technical Medical Radiology, University of Applied Sciences (HES-SO), HESAV, Lausanne, Switzerland. 2. Symbios Orthopédie SA, Yverdon-les-Bains, Switzerland. 3. Department of Technical Medical Radiology, University of Applied Sciences (HES-SO), HESAV, Lausanne, Switzerland. Electronic address: sandrine.ding@hesav.ch.
Abstract
OBJECTIVES: The purposes of this study were to (1) investigate the limits of measurements on scout view in three computed tomography axes, x, y and z and (2) develop a model to provide better understanding of measurement accuracy. METHODS: For the first objective, anteroposterior and lateral scout views of a Catphan phantom 200 mm in diameter and length were acquired with a GE scanner at 21 different table heights. Phantom measurements on scout view were performed by two experienced readers. The comparison of their measures provided estimation of precision. The accuracy was assessed by determining the bias, calculated as the difference between the values measured on scout view and the real phantom size. Second, a model was developed investigating the relationship between the dimensions of the object, its image, and the table height. This relationship was tested on our data. RESULTS: Scout view measurements were precise, with less than 0.53% difference between readers. In addition, small biases of about 1 mm were detected in the z-axis, whatever the table height. In the other axes, serious biases from -13 to +73 mm were measured. Furthermore, at isocentre, overestimations up to 7 mm were shown. The results also indicated that biases in scout view measurements are because of the geometrical projection related to the object-detector distance. CONCLUSIONS: Measurements in the table movement axis are precise and accurate, conferring to scout views an added value for preoperative planning in orthopedic surgery.
OBJECTIVES: The purposes of this study were to (1) investigate the limits of measurements on scout view in three computed tomography axes, x, y and z and (2) develop a model to provide better understanding of measurement accuracy. METHODS: For the first objective, anteroposterior and lateral scout views of a Catphan phantom 200 mm in diameter and length were acquired with a GE scanner at 21 different table heights. Phantom measurements on scout view were performed by two experienced readers. The comparison of their measures provided estimation of precision. The accuracy was assessed by determining the bias, calculated as the difference between the values measured on scout view and the real phantom size. Second, a model was developed investigating the relationship between the dimensions of the object, its image, and the table height. This relationship was tested on our data. RESULTS: Scout view measurements were precise, with less than 0.53% difference between readers. In addition, small biases of about 1 mm were detected in the z-axis, whatever the table height. In the other axes, serious biases from -13 to +73 mm were measured. Furthermore, at isocentre, overestimations up to 7 mm were shown. The results also indicated that biases in scout view measurements are because of the geometrical projection related to the object-detector distance. CONCLUSIONS: Measurements in the table movement axis are precise and accurate, conferring to scout views an added value for preoperative planning in orthopedic surgery.
Authors: Maximilian Nunninger; Victor Paul Bela Braun; Marco Ziegert; Felix Benjamin Schwarz; Bernd Hamm; Michael Scheel; Paul Jahnke Journal: Neuroradiology Date: 2019-12-14 Impact factor: 2.804