Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2022 Jun 14;8(3):1552-1569.
doi: 10.3390/tomography8030128.

The Influence of Data-Driven Compressed Sensing Reconstruction on Quantitative Pharmacokinetic Analysis in Breast DCE MRI

Ping Ni Wang  1 , Julia V Velikina  2 , Leah C Henze Bancroft  2 , Alexey A Samsonov  2 , Frederick Kelcz  2 , Roberta M Strigel  1   2   3 , James H Holmes  4   5
Affiliations

The Influence of Data-Driven Compressed Sensing Reconstruction on Quantitative Pharmacokinetic Analysis in Breast DCE MRI

Ping Ni Wang et al. Tomography. .

Abstract

Radial acquisition with MOCCO reconstruction has been previously proposed for high spatial and temporal resolution breast DCE imaging. In this work, we characterize MOCCO across a wide range of temporal contrast enhancement in a digital reference object (DRO). Time-resolved radial data was simulated using a DRO with lesions in different PK parameters. The under sampled data were reconstructed at 5 s temporal resolution using the data-driven low-rank temporal model for MOCCO, compressed sensing with temporal total variation (CS-TV) and more conventional low-rank reconstruction (PCB). Our results demonstrated that MOCCO was able to recover curves with Ktrans values ranging from 0.01 to 0.8 min-1 and fixed Ve = 0.3, where the fitted results are within a 10% bias error range. MOCCO reconstruction showed less impact on the selection of different temporal models than conventional low-rank reconstruction and the greater error was observed with PCB. CS-TV showed overall underestimation in both Ktrans and Ve. For the Monte-Carlo simulations, MOCCO was found to provide the most accurate reconstruction results for curves with intermediate lesion kinetics in the presence of noise. Initial in vivo experiences are reported in one patient volunteer. Overall, MOCCO was able to provide reconstructed time-series data that resulted in a more accurate measurement of PK parameters than PCB and CS-TV.

Keywords: breast DCE-MRI; compressed sensing; quantitative imaging.

PubMed Disclaimer

Conflict of interest statement

The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. The University of Wisconsin-Madison receives research support from GE Healthcare. Pingni Wang was an employee of the University of Wisconsin-Madison during this work and is now an employee of GE Healthcare.

Figures

Figure 1
Figure 1
(A) A breast digital reference object (DRO) (matrix size 448 × 448 × 142) phantom is shown with one lesion located in the fibroglandular tissue. (B) Nine configurations of the DRO phantom were evaluated, each with an 8 mm diameter lesion at the same location and having contrast kinetics generated by assigning the fixed Ve = 0.3, fixed Vp = 0.001, and Ktrans ranging from 0.01 to 1.5 using the extended Tofts model, respectively.
Figure 2
Figure 2
Simulated CTCs with slow (A,D,G), intermediate (B,E,H) and rapid (C,F,I) contrast kinetics in noise-free data (displayed for a subset of time from 150 to 400 s). Mean CTCs measured for three lesions with varying Ktrans values reconstructed using CS-TV (green star) (G–I), MOCCO (red star) and PCB (blue circle) with the temporal model derived from high spatial resolution (HR) (A–C) and low spatial resolution (LR) images (D–F). The corresponding standard deviations within the lesions are shown with banded area. The input time curves (“truth”) used to generate the source data are plotted with dark black lines in all frames.
Figure 3
Figure 3
The influence of temporal model in noise-free data using MOCCO (A,C), PCB (B,D) and CS-TV (E) reconstruction on parameter estimation of Ktrans and Ve. Bland–Altman plots show the mean (±standard deviation) of Ktrans (blue stars) and Ve (red circles). The ±10% error range is shown as black dashed lines. Note the results for the HR approaches (A,B) represent idealized scenarios where the full spatial resolution temporal model can be utilized whereas the LR approaches (C,D) represent more realistic scenarios where the temporal model could be learned from the under sampled data.
Figure 4
Figure 4
The influence of temporal model and noise using MOCCO (A,C), PCB (B,D) and CS-TV (E) reconstruction on parameter estimation of Ktrans and Ve. Bland–Altman plots show the mean (±standard deviation) of Ktrans (blue stars) and Ve (red circles). The ±10% error range is shown as black dash lines.
Figure 5
Figure 5
Visualization of zoomed-in error maps for Ktrans and Ve from (A) MOCCO-HR, (B) PCB-HR, (C) MOCCO-LR, (D) PCB-LR and (E) CS-TV without noise added to the simulated lesions with Ktrans = 0.01, 0.3, 1.5 min−1, obtained by measuring the % differences between the fitted parameters and the true values for the lesion.
Figure 6
Figure 6
Visualization of zoomed-in error maps for Ktrans and Ve from (A) MOCCO-HR, (B) PCB-HR, (C) MOCCO-LR, (D) PCB-LR and (E) CS-TV with 20% noise added to the simulated lesions with Ktrans = 0.01, 0.3, 1.5 min−1, obtained by measuring the % differences between the fitted parameters and the true values for the lesion.
Figure 7
Figure 7
Simulated CTCs with 20% noise added (displayed for a subset of time from 150 s to 400 s). Mean CTCs measured for three lesions with Ktrans values of 0.1 min−1 (A), 0.3 min−1 (B) and 1.5 min−1 (C). values reconstructed using MOCCO-LR. The input time curves (“truth”) used to generate the source data are plotted with dark black lines in all frames.
Figure 8
Figure 8
Visualization of zoomed-in color maps for Ktrans and Ve from fully-sampled data with 20% noise added to Ktrans = 0.01, 0.3, 1.5 min−1 obtained by measuring the (A) mean, (B) standard deviation, (C) percent differences between the fitted parameters from all Monte-Carlo noise realizations and the true values for the lesions.
Figure 9
Figure 9
Visualization of zoomed-in color maps for Ktrans and Ve from MOCCO-LR with 20% noise added to Ktrans = 0.01, 0.3, 1.5 min−1 obtained by measuring the (A) mean, (B) standard deviation, (C) percent differences between the fitted parameters from all Monte-Carlo noise realizations and the true values for the lesions.
Figure 10
Figure 10
Time-resolved DCE images from a patient volunteer reconstructed using MOCCO-LR with 5 s temporal resolution (A). Curves of the percent signal change (PSC) are plotted from ROIs placed in the lesion ((B), blue), muscle ((C), green), aorta ((D), red) and lymph node ((E), yellow).
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Mann R.M., Cho N., Moy L. Breast MRI: State of the Art. Radiology. 2019;292:520–536. doi: 10.1148/radiol.2019182947. - DOI - PubMed
    1. Pinker-Domenig K., Bogner W., Gruber S., Bickel H., Duffy S., Schernthaner M., Dubsky P., Pluschnig U., Rudas M., Trattnig S., et al. High Resolution MRI of the Breast at 3 T: Which BI-RADS® Descriptors Are Most Strongly Associated with the Diagnosis of Breast Cancer? Eur. Radiol. 2012;22:322–330. doi: 10.1007/s00330-011-2256-6. - DOI - PubMed
    1. Harry V.N., Semple S.I., Parkin D.E., Gilbert F.J. Use of New Imaging Techniques to Predict Tumour Response to Therapy. Lancet Oncol. 2010;11:92–102. doi: 10.1016/S1470-2045(09)70190-1. - DOI - PubMed
    1. Lobbes M.B.I., Prevos R., Smidt M., Tjan-Heijnen V.C.G., van Goethem M., Schipper R., Beets-Tan R.G., Wildberger J.E. The Role of Magnetic Resonance Imaging in Assessing Residual Disease and Pathologic Complete Response in Breast Cancer Patients Receiving Neoadjuvant Chemotherapy: A Systematic Review. Insights Imaging. 2013;4:163–175. doi: 10.1007/s13244-013-0219-y. - DOI - PMC - PubMed
    1. Li X., Arlinghaus L.R., Ayers G.D., Chakravarthy A.B., Abramson R.G., Abramson V.G., Atuegwu N., Farley J., Mayer I.A., Kelley M.C., et al. DCE-MRI Analysis Methods for Predicting the Response of Breast Cancer to Neoadjuvant Chemotherapy: Pilot Study Findings. Magn. Reason. Med. 2014;71:1592–1602. doi: 10.1002/mrm.24782. - DOI - PMC - PubMed

Publication types