Premium
WE‐C‐103‐12: Individually Optimized Contrast Injection in CT Angiography for the Diagnosis of Pulmonary Thromboembolic Disease
Author(s) -
Xue M,
Zhang H,
Kligerman S,
D' Souza W,
Klahr P,
Lu W
Publication year - 2013
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 - 10.1118/1.4815561
Subject(s) - contrast (vision) , mean squared error , angiography , mathematics , nuclear medicine , medicine , radiology , computer science , artificial intelligence , statistics
Purpose: To improve the diagnostic quality of CT angiography (CTPA) for pulmonary thromboembolic disease by individually optimizing a bi‐phasic contrast injection function to achieve uniform target contrast enhancement. To compare the results against a previously reported discrete Fourier transform (DFT) approach. Methods: This simulation study uses a retrospective dataset of 27 consecutive patients. We developed an optimization approach consists of two steps: 1. Compute the impulse enhancement function (IEF) based on the test bolus scan; 2. Optimize the contrast injection function using the IEF in order to achieve uniform target enhancement. We chose a bi‐phasic contrast injection function in which the injection rates and durations are optimized by minimizing the difference between its contrast enhancement curve and the uniform target enhancement curve. The contrast volume is limited either to the clinical standard of 65 mL or to the same amount used in the DFT approach. The optimization approach is compared against the DFT approaches in terms of the root mean square errors (RMSE) from the uniform target contrast enhancement curve and the contrast volume used. Results: When the contrast volume is limited to 65 mL, the optimization approach produces contrast enhancement significantly (p<.00001) closer to the uniform target contrast enhancement (mean RMSE 17 HU) than the DFT approach (mean RMSE 56 HU), for which the contrast volume is not even limited. When the contrast volume is limited to be the same amount used in the DFT approach, the optimization approach still generate significantly improved contrast enhancement than the DFT approach (RMSE 44 HU vs. 56 HU, p<.0099). Conclusion: The optimization approach generates individually optimized bi‐phasic injection functions yielding significantly improved contrast enhancement than the DFT approach. Its implementation in clinic has great potential to improve the diagnostic quality of CTPA. This work is supported in part by Philips Healthcare, Inc.