Deep Learning-Based Generation of DSC MRI Parameter Maps Using Dynamic Contrast-Enhanced MRI Data.
Authors
Affiliations (12)
Affiliations (12)
- From the BioMedical Engineering and Imaging Institute (H.P., L.F., X.X), Icahn School of Medicine at Mount Sinai, New York, New York.
- Center for Advanced Imaging Innovation and Research (CAIR) (H.P., L.F.), New York University Grossman School of Medicine, New York, New York.
- Tandon School of Engineering (H.P., Y.L.), New York University, New York, New York.
- Image Processing Center, School of Astronautics (Y.L.), Beihang University, Beijing, China.
- Department of Neurology (S.L.), Technical University of Munich, Munich, Germany.
- Department of Radiology (S.L., D.L., P.v.Z., L.K., X.X.), Johns Hopkins University, Baltimore, Maryland.
- Institute of Neuroradiology (S.L.), Ludwig-Maximilians-Universität, Munich, Germany.
- Athinoula A. Martinos Center for Biomedical Imaging (F.L.), Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts.
- Department of Neurology (P.N.), Johns Hopkins University, Baltimore, Maryland.
- F.M. Kirby Research Center for Functional Brain Imaging (P.v.Z., L.K.), Kennedy Krieger Institute, Baltimore, Maryland.
- Department of Medical Radiation Physics (L.K.), Lund University, Lund, Sweden.
- From the BioMedical Engineering and Imaging Institute (H.P., L.F., X.X), Icahn School of Medicine at Mount Sinai, New York, New York [email protected].
Abstract
Perfusion and perfusion-related parameter maps obtained by using DSC MRI and dynamic contrast-enhanced (DCE) MRI are both useful for clinical diagnosis and research. However, using both DSC and DCE MRI in the same scan session requires 2 doses of gadolinium contrast agent. The objective was to develop deep learning-based methods to synthesize DSC-derived parameter maps from DCE MRI data. Independent analysis of data collected in previous studies was performed. The database contained 64 participants, including patients with and without brain tumors. The reference parameter maps were measured from DSC MRI performed after DCE MRI. A conditional generative adversarial network (cGAN) was designed and trained to generate synthetic DSC-derived maps from DCE MRI data. The median parameter values and distributions between synthetic and real maps were compared by using linear regression and Bland-Altman plots. Using cGAN, realistic DSC parameter maps could be synthesized from DCE MRI data. For controls without brain tumors, the synthesized parameters had distributions similar to the ground truth values. For patients with brain tumors, the synthesized parameters in the tumor region correlated linearly with the ground truth values. In addition, areas not visible due to susceptibility artifacts in real DSC maps could be visualized by using DCE-derived DSC maps. DSC-derived parameter maps could be synthesized by using DCE MRI data, including susceptibility-artifact-prone regions. This shows the potential to obtain both DSC and DCE parameter maps from DCE MRI by using a single dose of contrast agent.