Victoria Y Yu^{1}, Kathryn R Tringale^{2}, Ricardo Otazo^{1}, and Ouri Cohen^{1}

^{1}Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, United States, ^{2}Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, United States

In MR fingerprinting, quantitative maps are obtained by matching the measured signal to a pre-computed dictionary. However, a key constraint of dictionary matching is the exponential growth of the dictionary with the number of parameters. A deep learning method named DRONE overcomes this constraint by using deep learning to map the magnitude-valued signal to the underlying tissue parameters. Here we describe an extension of DRONE that jointly estimates a phase term to enable mapping complex-valued signals and improve the quantitative accuracy. We test the accuracy in the ISMRM NIST phantom and demonstrate the clinical utility in patients with brain metastases.

Images were acquired with an EPI based MRF pulse sequence (MRF-EPI) whose 50-point schedule of flip angles (FA) and repetition times (TR) was optimized to maximize tissue discrimination[6]. The remaining acquisition parameters were as follows: partial Fourier factor of ~6/8, acceleration factor R=3, echo time=24 ms, matrix size=224×224, FOV=280 mm2 for an in-plane resolution of 1.25 mm2 and a slice thickness of 5 mm. The scan time was 5 seconds per slice.

The outline of the PS-DRONE method is shown in Figure 1. Like its predecessor, PS-DRONE uses a training dataset of signal magnetizations generated by simulating an MRF acquisition for different tissue parameter values. In PS-DRONE, however, each signal includes a multiplicative phase term exp(jΦ) that accounts for phase variations in the signal. In the network inference stage, the value of Φ for each voxel is estimated along with the other parameters (i.e. T1, T2 and B1) In this work, the network was trained with a 400,000 entries dictionary selected from the following ranges: T1=[1, 4000], T2=[1, 3000], B1=[0, 1.5], Φ=[-π, π]. The proton density (PD) was calculated as a scaling factor from the reconstructed data.

All experiments were conducted on a Signa Premier 3T scanner (GE Healthcare, Waukesha, WI) with a 48-channel head receiver coil. The ISMRM NIST phantom was scanned with the MRF-EPI sequence and the T1 and T2 maps quantified with PS-DRONE. Regions-of-interest were drawn around each compartment and the mean and standard deviation calculated and compared to the reference values.

A healthy, 30 years old female volunteer was recruited and gave informed consent in accordance with the institutional IRB protocol. The subject was scanned with the MRF-EPI sequence and the data reconstructed with PS-DRONE as described above. For comparison, we also reconstructed the data using conventional DRONE using the magnitude images and the complex data but without the phase estimation.

Three subjects with metastatic brain tumors were recruited for this study and gave informed consent. The tissue maps were quantified with PS-DRONE and each lesion segmented into tumor, necrotic and edema regions by a trained radiation oncologist. A healthy contra-lateral region was also demarcated. The mean and standard deviation of the tissue parameter values in each region were compared across patients and compared to values obtained with a standard-of-care protocol in the same scan session for each patient.

[1] D. Ma et al., “Magnetic resonance fingerprinting,” Nature, vol. 495, no. 7440, pp. 187–192, 2013.

[2] O. Cohen, B. Zhu, and M. S. Rosen, “MR fingerprinting deep reconstruction network (DRONE),” Magn. Reson. Med., vol. 80, no. 3, pp. 885–894, 2018.

[3] N. Stikov, M. Boudreau, I. R. Levesque, C. L. Tardif, J. K. Barral, and G. B. Pike, “On the accuracy of T1 mapping: searching for common ground,” Magn. Reson. Med., vol. 73, no. 2, pp. 514–522, 2015.

[4] P. Virtue, X. Y. Stella, and M. Lustig, “Better than real: Complex-valued neural nets for MRI fingerprinting,” in 2017 IEEE international conference on image processing (ICIP), 2017, pp. 3953–3957.

[5] K. F. Stupic et al., “A standard system phantom for magnetic resonance imaging,” Magn. Reson. Med., 2021.

[6] O. Cohen and M. S. Rosen, “Algorithm comparison for schedule optimization in MR fingerprinting,” Magn. Reson. Imaging, vol. 41, pp. 15–21, 2017.

[7] Y. Wang and T. Liu, “Quantitative susceptibility mapping (QSM): decoding MRI data for a tissue magnetic biomarker,” Magn. Reson. Med., vol. 73, no. 1, pp. 82–101, 2015.

Figure 1: Diagram of the PS-DRONE method. The network is
trained with a dataset simulated for different values of the tissue parameters
which includes a multiplicative phase term Φ. In the inference stage, the real
and imaginary measured data is fed into the network which outputs the
corresponding parameters and the estimated phase for each voxel.

Figure 2: Estimated mean and standard deviation T1 and T2
values in the ISMRM NIST phantom in comparison to the reference values. The
dashed line is the identity line. Note the excellent agreement between the
reference and estimated values.

Figure 3: A comparison of different DRONE quantifications.
(A-D) Quantitative maps obtained using magnitude only images resulted in errors
in the T1 and T2 maps. (E-H) Reconstruction with the complex data but without
accounting for the phase resulted in severe errors and artifacts. (I-M)
Reconstruction using the complex data and inclusion of the phase term Φ yielded
the error-free results.

Figure 4: Quantitative tissue maps obtained with PS-DRONE
from a 41 years old male subject with metastatic melanoma. A subset of lesions
is indicated by the red arrows with edema indicated by yellow arrows.

Figure 5: (A) Scatter plot of the MRF derived T1, T2 and
proton density values in each of the segmented tumor regions. (B) Standard-of-care
imaging sequence (ADC, FLAIR and post-contrast T1) values for the same regions.
(C) Projection of the results in (A) onto the T1-T2 plane for easy
visualization, (D) Projection of the results in (B) onto the ADC-FLAIR plane. Note
that the marker size indicates the original primary tumor.

DOI: https://doi.org/10.58530/2022/0177