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

Title: A method for patient dose reduction in dynamic contrast enhanced CT study

Abstract

Purpose: In dynamic contrast enhanced CT (DCE-CT) study, prolonged CT scanning with high temporal resolution is required to give accurate and precise estimates of kinetic parameters. However, such scanning protocol could lead to substantial radiation dose to the patient. A novel method is proposed to reduce radiation dose to patient, while maintaining high accuracy for kinetic parameter estimates in DCE-CT study. Methods: The method is based on a previous investigation that the arterial impulse response (AIR) in DCE-CT study can be predicted using a population-based scheme. In the proposed method, DCE-CT scanning is performed with relatively low temporal resolution, hence, giving rise to reduction in patient dose. A novel method is proposed to estimate the arterial input function (AIF) based on the coarsely sampled AIF. By using the estimated AIF in the tracer kinetic analysis of the coarsely sampled DCE-CT study, the calculated kinetic parameters are able to achieve a high degree of accuracy. The method was tested on a DCE-CT data set of 48 patients with cervical cancer scanned at high temporal resolution. A random cohort of 34 patients was chosen to construct the orthonormal bases of the AIRs via singular value decomposition method. The determined set of orthonormalmore » bases was used to fit the AIFs in the second cohort (14 patients) at varying levels of down sampling. For each dataset in the second cohort, the estimated AIF was used for kinetic analyses of the modified Tofts and adiabatic tissue homogeneity models for each of the down-sampling schemes between intervals from 2 to 15 s. The results were compared with analyses done with the ''raw'' down-sampled AIF. Results: In the first group of 34 patients, there were 11 orthonormal bases identified to describe the AIRs. The AIFs in the second group were estimated in high accuracy based on the 11 orthonormal bases established in the first group along with down-sampled AIFs. Using the 11 orthonormal bases, the estimated AIFs for the second group were found to have an averaged maximal percentage error of 3.4%{+-} 7.5% in all sampling schemes up to 15 s. The results of kinetic analysis with the proposed method compared with down sampling alone showed that the proposed method is superior in maintaining the accuracy in volume transfer constant (K{sup trans} ) after 9 s down-sampling interval, blood volume (v{sub b} ) for almost all down-sampling intervals, and blood flow (F) after 11 s down-sampling interval. The preliminary results suggested that the proposed method is able to support scanning intervals of 10-15 s at a cost of 6.2%-10.0% loss in accuracy of K{sup trans} and 10.9%-19.4% in v{sub b}, and the scanning intervals of 12-15 s at a cost of 9.7%-14.6% for F in DEC-CT studies for patients with cervix cancer. Conclusions: The proposed method of AIF estimation allows low scanning frequency in DCE-CT study to reduce radiation dose to patient, while maintaining relatively high accuracy in the kinetic parameter estimates. The initial results suggested that the method is applicable for DCE-CT studies for patients with cervical cancer.« less

Authors:
; ; ; ;  [1];  [2];  [2]
  1. Radiation Medicine Program, Princess Margaret Hospital/University Health Network, Toronto, Ontario, M5G 2M9 (Canada)
  2. (Canada)
Publication Date:
OSTI Identifier:
22100605
Resource Type:
Journal Article
Journal Name:
Medical Physics
Additional Journal Information:
Journal Volume: 38; Journal Issue: 9; Other Information: (c) 2011 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:
61 RADIATION PROTECTION AND DOSIMETRY; 60 APPLIED LIFE SCIENCES; ACCURACY; BLOOD FLOW; CAT SCANNING; COMPUTERIZED TOMOGRAPHY; DATASETS; DECOMPOSITION; DOSIMETRY; ERRORS; IMAGE PROCESSING; LOSSES; NEOPLASMS; PULSES; RADIATION DOSES; RANDOMNESS; RESOLUTION; SAMPLING

Citation Formats

Mo Kim, Sun, Haider, Masoom A., Milosevic, Michael, Jaffray, David A., Yeung, Ivan W. T., Department of Medical Imaging, Princess Margaret Hospital/University Health Network, Toronto, Ontario, M5G 2M9, and Radiation Medicine Program, Princess Margaret Hospital/University Health Network, Toronto, Ontario, M5G 2M9, Canada and Department of Radiation Oncology, University of Toronto, Ontario, M5G 2M9. A method for patient dose reduction in dynamic contrast enhanced CT study. United States: N. p., 2011. Web. doi:10.1118/1.3622611.
Mo Kim, Sun, Haider, Masoom A., Milosevic, Michael, Jaffray, David A., Yeung, Ivan W. T., Department of Medical Imaging, Princess Margaret Hospital/University Health Network, Toronto, Ontario, M5G 2M9, & Radiation Medicine Program, Princess Margaret Hospital/University Health Network, Toronto, Ontario, M5G 2M9, Canada and Department of Radiation Oncology, University of Toronto, Ontario, M5G 2M9. A method for patient dose reduction in dynamic contrast enhanced CT study. United States. doi:10.1118/1.3622611.
Mo Kim, Sun, Haider, Masoom A., Milosevic, Michael, Jaffray, David A., Yeung, Ivan W. T., Department of Medical Imaging, Princess Margaret Hospital/University Health Network, Toronto, Ontario, M5G 2M9, and Radiation Medicine Program, Princess Margaret Hospital/University Health Network, Toronto, Ontario, M5G 2M9, Canada and Department of Radiation Oncology, University of Toronto, Ontario, M5G 2M9. Thu . "A method for patient dose reduction in dynamic contrast enhanced CT study". United States. doi:10.1118/1.3622611.
@article{osti_22100605,
title = {A method for patient dose reduction in dynamic contrast enhanced CT study},
author = {Mo Kim, Sun and Haider, Masoom A. and Milosevic, Michael and Jaffray, David A. and Yeung, Ivan W. T. and Department of Medical Imaging, Princess Margaret Hospital/University Health Network, Toronto, Ontario, M5G 2M9 and Radiation Medicine Program, Princess Margaret Hospital/University Health Network, Toronto, Ontario, M5G 2M9, Canada and Department of Radiation Oncology, University of Toronto, Ontario, M5G 2M9},
abstractNote = {Purpose: In dynamic contrast enhanced CT (DCE-CT) study, prolonged CT scanning with high temporal resolution is required to give accurate and precise estimates of kinetic parameters. However, such scanning protocol could lead to substantial radiation dose to the patient. A novel method is proposed to reduce radiation dose to patient, while maintaining high accuracy for kinetic parameter estimates in DCE-CT study. Methods: The method is based on a previous investigation that the arterial impulse response (AIR) in DCE-CT study can be predicted using a population-based scheme. In the proposed method, DCE-CT scanning is performed with relatively low temporal resolution, hence, giving rise to reduction in patient dose. A novel method is proposed to estimate the arterial input function (AIF) based on the coarsely sampled AIF. By using the estimated AIF in the tracer kinetic analysis of the coarsely sampled DCE-CT study, the calculated kinetic parameters are able to achieve a high degree of accuracy. The method was tested on a DCE-CT data set of 48 patients with cervical cancer scanned at high temporal resolution. A random cohort of 34 patients was chosen to construct the orthonormal bases of the AIRs via singular value decomposition method. The determined set of orthonormal bases was used to fit the AIFs in the second cohort (14 patients) at varying levels of down sampling. For each dataset in the second cohort, the estimated AIF was used for kinetic analyses of the modified Tofts and adiabatic tissue homogeneity models for each of the down-sampling schemes between intervals from 2 to 15 s. The results were compared with analyses done with the ''raw'' down-sampled AIF. Results: In the first group of 34 patients, there were 11 orthonormal bases identified to describe the AIRs. The AIFs in the second group were estimated in high accuracy based on the 11 orthonormal bases established in the first group along with down-sampled AIFs. Using the 11 orthonormal bases, the estimated AIFs for the second group were found to have an averaged maximal percentage error of 3.4%{+-} 7.5% in all sampling schemes up to 15 s. The results of kinetic analysis with the proposed method compared with down sampling alone showed that the proposed method is superior in maintaining the accuracy in volume transfer constant (K{sup trans} ) after 9 s down-sampling interval, blood volume (v{sub b} ) for almost all down-sampling intervals, and blood flow (F) after 11 s down-sampling interval. The preliminary results suggested that the proposed method is able to support scanning intervals of 10-15 s at a cost of 6.2%-10.0% loss in accuracy of K{sup trans} and 10.9%-19.4% in v{sub b}, and the scanning intervals of 12-15 s at a cost of 9.7%-14.6% for F in DEC-CT studies for patients with cervix cancer. Conclusions: The proposed method of AIF estimation allows low scanning frequency in DCE-CT study to reduce radiation dose to patient, while maintaining relatively high accuracy in the kinetic parameter estimates. The initial results suggested that the method is applicable for DCE-CT studies for patients with cervical cancer.},
doi = {10.1118/1.3622611},
journal = {Medical Physics},
issn = {0094-2405},
number = 9,
volume = 38,
place = {United States},
year = {2011},
month = {9}
}