Premium
WE‐G‐18C‐03: Effects of Breathing Irregularity On Tumor Motion Measurement of 4D‐MRI: A Comparison Between Cine and Sequential Image Acquisition Modes
Author(s) -
Liu Y,
Yin F,
Rhee D,
Cai J
Publication year - 2014
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 - 10.1118/1.4889522
Subject(s) - imaging phantom , torso , nuclear medicine , breathing , medical imaging , mathematics , artificial intelligence , computer science , medicine , anatomy
Purpose: We have recently developed a cine‐mode T2*/T1‐weighted 4DMRI technique, and a sequential‐mode T2‐weighted 4D‐MRI for imaging respiratory motion . The latter has better tumor‐contrast due to its T2‐ weighting. This study aims at investigating which 4D‐MRI image acquisition mode, cine or sequential, provides more accurate measurement of respiratory motion. Methods: A 4D Digital Extended Cardiac‐Torso (XCAT) human phantom with a hypothesized tumor was used to simulate image acquisition and 4D‐MRI reconstruction. Tumor was set to move continuously with a given breathing signal. Its trajectories were measured from both sequential‐ and cine‐mode 4D‐MRI. Comparisons of measurements with the average tumor trajectories calculated from the input profile as references were conducted. Absolute tumor motion amplitude differences (D) were determined. A total of 500 simulated respiratory profiles with a wide range of irregularity were used to investigate the relationship between D and Ir. Statistical analysis of D for breathing profiles of 20 real cancer patients were conducted with a sign rank test regarding two modes. Lastly, we investigated the possibility of further improving motion measurement accuracy in sequential‐mode 4D‐MRI by removing data points of high irregularity. An irregularity filter was applied to the same 20 patients. Sign rank tests for D of sequential‐mode with filter and cine‐mode were performed. Results: D increased faster for cine‐mode (D=0.37*Ir) than sequential‐mode (D=0.16*Ir) as irregularity increased. For 20 cancer patients, D was 0.12cm and 0.10cm for cine‐ and sequentialmodes on average; p‐value was 0.0228. Removing highly irregular data points increased accuracy of tumor trajectory for sequential‐mode 4D‐MRI, D was decreased by 10%. P‐value was 0.0065. Conclusion: Tumor motion measurement is more accurate and less susceptible to breathing irregularity in sequential‐mode 4D‐MRI than that in cine‐mode 4D‐MRI; its accuracy could be further improved in sequential‐mode 4D‐MRI by selectively removing data points of high irregularity. NIH (1R21CA165384‐01A1)