Kalina P Slavkova^{1}, Julie C DiCarlo^{2}, Anum K Syed^{3}, John Virostko^{4}, Anna G Sorace^{4}, and Thomas E Yankeelov^{2,3,4}

MRI is the most sensitive imaging modality for detecting breast cancer, but its use as a screening tool is limited. There has been recent interest in developing an “abbreviated” breast MRI protocol as a screening protocol, which involves a significantly shorter breast MRI exam that does not compromise diagnostic accuracy. In this work, we evaluate the accuracy of estimating perfusion parameters from dynamic contrast-enhanced MRI data by truncating the data into a series of abbreviated-time courses and comparing the corresponding parameter estimates to the original, full-time course parameter estimates.

DCE-MRI simulation: Simulated data was produced using a MATLAB (Mathworks, Natick, MA) implementation of the standard Kety-Tofts model^{4}. We simulated signal intensity curves (N=729) using combinations of *K ^{trans }*and

DCE-MRI data acquisition: Patients (N=22) with locally advanced breast cancer were scanned using a 3T Skyra (Siemens, Tarrytown, NY) equipped with a 16-channel receive double-breast coil (Invivo, Gainsville, FL). DCE-MRI data was collected with *TR/TE/a *= 7.02 ms/4.60 ms/6^{o} and a GRAPPA acceleration factor of 2 so that each 10-slice set was collected^{} in 7.27 s for eight total minutes of scanning, yielding 66 total time points. After collecting one minute of dynamic scans (first 8 time points), 10 mL of Gadavist (Bayer, Whippany, NJ) was delivered at 2 mL/sec (followed by a saline fush) through a catheter placed within an antecubital vein. One representative patient dataset was used for the preliminary analysis presented in this work.

DCE-MRI analysis: Two sets of seven abbreviated-time courses (ATCs) of the same temporal resolution with respective lengths of 16 (1.94 min), 24 (2.91 min), 32 (3.88 min), 40 (4.85 min), 48 (5.82 min), 56 (6.79 min), and 64 (7.75 min) time points were generated by truncating the full-time course (FTC; 66 time points, 8 min length) from the patient and the simulated data. The FTCs for both the patient and simulated data were fit to the standard Kety-Tofts model to estimate *K ^{trans}* and

Statistical analysis: We calculated the average percent error between the FTC and ATC estimates of *K ^{trans} *and

Results from simulated data are contained in **Figure 1**, which summarizes the average percent error in *K ^{trans} *and

Results from patient data are contained in **Figure 2**, which summarizes the average percent error in *K ^{trans}* and

- Leithner D, Moy L, et al. Abbreviated MRI of the Breast: Does It Provide Value? J. Magn. Reson. Imaging. 2018
- Amarnath J, Sangeeta T, Mehta SB. Role of quantitative pharmacokinetic parameter (transfer constant: K(trans)) in the characterization of breast lesions on MRI. Indian J Radiol Imaging. 2013;23(1):19-25.
- Levine E, Daniel B, et al. 3D Cartesian MRI with Compressed Sensing and Variable View Sharing Using Complementary Poisson-disc Sampling. Magn Reson Med. 2017;77(5):1774-1785.
- Tofts P, Kermode A. Measurement of the Blood-Brain Barrier Permeability and Leakage Space Using Dynamic MR Imaging. 1. Fundamental Concepts. Magn. Reson. in Med. 1991;17:357-367.