skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: A dual-Kinect approach to determine torso surface motion for respiratory motion correction in PET

Abstract

Purpose: Respiratory gating is commonly used to reduce blurring effects and attenuation correction artifacts in positron emission tomography (PET). Established clinically available methods that employ body-attached hardware for acquiring respiration signals rely on the assumption that external surface motion and internal organ motion are well correlated. In this paper, the authors present a markerless method comprising two Microsoft Kinects for determining the motion on the whole torso surface and aim to demonstrate its validity and usefulness—including the potential to study the external/internal correlation and to provide useful information for more advanced correction approaches. Methods: The data of two Kinects are used to calculate 3D representations of a patient’s torso surface with high spatial coverage. Motion signals can be obtained for any position by tracking the mean distance to a virtual camera with a view perpendicular to the surrounding surface. The authors have conducted validation experiments including volunteers and a moving high-precision platform to verify the method’s suitability for providing meaningful data. In addition, the authors employed it during clinical {sup 18}F-FDG-PET scans and exemplarily analyzed the acquired data of ten cancer patients. External signals of abdominal and thoracic regions as well as data-driven signals were used for gating and comparedmore » with respect to detected displacement of present lesions. Additionally, the authors quantified signal similarities and time shifts by analyzing cross-correlation sequences. Results: The authors’ results suggest a Kinect depth resolution of approximately 1 mm at 75 cm distance. Accordingly, valid signals could be obtained for surface movements with small amplitudes in the range of only few millimeters. In this small sample of ten patients, the abdominal signals were better suited for gating the PET data than the thoracic signals and the correlation of data-driven signals was found to be stronger with abdominal signals than with thoracic signals (average Pearson correlation coefficients of 0.74 ± 0.17 and 0.45 ± 0.23, respectively). In all cases, except one, the abdominal respiratory motion preceded the thoracic motion—a maximum delay of approximately 600 ms was detected. Conclusions: The method provides motion information with sufficiently high spatial and temporal resolution. Thus, it enables meaningful analysis in the form of comparisons between amplitudes and phase shifts of signals from different regions. In combination with a large field-of-view, as given by combining the data of two Kinect cameras, it yields surface representations that might be useful in the context of motion correction and motion modeling.« less

Authors:
; ; ;  [1];  [2]
  1. European Institute for Molecular Imaging, University of Münster, Münster 48149 (Germany)
  2. European Institute for Molecular Imaging, University of Münster, Münster 48149, Germany and Department of Mathematics and Computer Science, University of Münster, Münster 48149 (Germany)
Publication Date:
OSTI Identifier:
22413546
Resource Type:
Journal Article
Journal Name:
Medical Physics
Additional Journal Information:
Journal Volume: 42; Journal Issue: 5; Other Information: (c) 2015 American Association of Physicists in Medicine; Country of input: International Atomic Energy Agency (IAEA); Journal ID: ISSN 0094-2405
Country of Publication:
United States
Language:
English
Subject:
62 RADIOLOGY AND NUCLEAR MEDICINE; 60 APPLIED LIFE SCIENCES; ACCURACY; ATTENUATION; CORRECTIONS; FLUORINE 18; PATIENTS; POSITRON COMPUTED TOMOGRAPHY; RESPIRATION

Citation Formats

Heß, Mirco, E-mail: mirco.hess@uni-muenster.de, Büther, Florian, Dawood, Mohammad, Schäfers, Klaus P., and Gigengack, Fabian. A dual-Kinect approach to determine torso surface motion for respiratory motion correction in PET. United States: N. p., 2015. Web. doi:10.1118/1.4917163.
Heß, Mirco, E-mail: mirco.hess@uni-muenster.de, Büther, Florian, Dawood, Mohammad, Schäfers, Klaus P., & Gigengack, Fabian. A dual-Kinect approach to determine torso surface motion for respiratory motion correction in PET. United States. doi:10.1118/1.4917163.
Heß, Mirco, E-mail: mirco.hess@uni-muenster.de, Büther, Florian, Dawood, Mohammad, Schäfers, Klaus P., and Gigengack, Fabian. Fri . "A dual-Kinect approach to determine torso surface motion for respiratory motion correction in PET". United States. doi:10.1118/1.4917163.
@article{osti_22413546,
title = {A dual-Kinect approach to determine torso surface motion for respiratory motion correction in PET},
author = {Heß, Mirco, E-mail: mirco.hess@uni-muenster.de and Büther, Florian and Dawood, Mohammad and Schäfers, Klaus P. and Gigengack, Fabian},
abstractNote = {Purpose: Respiratory gating is commonly used to reduce blurring effects and attenuation correction artifacts in positron emission tomography (PET). Established clinically available methods that employ body-attached hardware for acquiring respiration signals rely on the assumption that external surface motion and internal organ motion are well correlated. In this paper, the authors present a markerless method comprising two Microsoft Kinects for determining the motion on the whole torso surface and aim to demonstrate its validity and usefulness—including the potential to study the external/internal correlation and to provide useful information for more advanced correction approaches. Methods: The data of two Kinects are used to calculate 3D representations of a patient’s torso surface with high spatial coverage. Motion signals can be obtained for any position by tracking the mean distance to a virtual camera with a view perpendicular to the surrounding surface. The authors have conducted validation experiments including volunteers and a moving high-precision platform to verify the method’s suitability for providing meaningful data. In addition, the authors employed it during clinical {sup 18}F-FDG-PET scans and exemplarily analyzed the acquired data of ten cancer patients. External signals of abdominal and thoracic regions as well as data-driven signals were used for gating and compared with respect to detected displacement of present lesions. Additionally, the authors quantified signal similarities and time shifts by analyzing cross-correlation sequences. Results: The authors’ results suggest a Kinect depth resolution of approximately 1 mm at 75 cm distance. Accordingly, valid signals could be obtained for surface movements with small amplitudes in the range of only few millimeters. In this small sample of ten patients, the abdominal signals were better suited for gating the PET data than the thoracic signals and the correlation of data-driven signals was found to be stronger with abdominal signals than with thoracic signals (average Pearson correlation coefficients of 0.74 ± 0.17 and 0.45 ± 0.23, respectively). In all cases, except one, the abdominal respiratory motion preceded the thoracic motion—a maximum delay of approximately 600 ms was detected. Conclusions: The method provides motion information with sufficiently high spatial and temporal resolution. Thus, it enables meaningful analysis in the form of comparisons between amplitudes and phase shifts of signals from different regions. In combination with a large field-of-view, as given by combining the data of two Kinect cameras, it yields surface representations that might be useful in the context of motion correction and motion modeling.},
doi = {10.1118/1.4917163},
journal = {Medical Physics},
issn = {0094-2405},
number = 5,
volume = 42,
place = {United States},
year = {2015},
month = {5}
}