Authors :
Presenting Author: Maksim Parfyonov, MD – Cleveland Clinic
Ting-Yu Su, MS – Epilepsy Center, Neurological Institute, Cleveland Clinic
Joon Yul Choi, PhD – Biomedical Engineering, Yonsei University, Wonju, Republic of Korea
Ingmar Blumcke, MD – Neuropathology, University Hospital Erlangen, Erlangen, Germany
Ken Sakaie, PhD – Imaging Institute, Cleveland Clinic
Andreas V. Alexopoulos, MD – Epilepsy Center, Neurological Institute, Cleveland Clinic
Imad Najm, MD – Cleveland Clinic
Dan Ma, PhD – Biomedical Engineering, Case Western Reserve University
Stephen E. Jones, MD, PhD – Imaging Institute, Cleveland Clinic
Zhong Irene Wang, PhD – Epilepsy Center, Neurological Institute, Cleveland Clinic
Rationale:
Magnetic resonance fingerprinting (MRF) is a novel paradigm of magnetic resonance imaging which uses pseudorandom acquisition parameters for radiofrequency flip angles and repetition times to yield quantitative signals which are then matched to a dictionary of known signal evolutions unique to various tissues (Ma et al., 2013). In contrast to conventional MRI, MRF is able to measure MR parameters of interest in a reproducible and quantitative manner.
Interest in MRF for epilepsy has grown in recent years, as a tool for studying epileptogenicity of periventricular nodular heterotopias (Choi et al., 2022), investigating group-level differences in epilepsy patients (Su et al., 2023; Tang et al., 2022), and as an input feature for machine learning models to predict epilepsy laterality(Tang et al., 2022) or lesion detection (Ding et al., 2024). Compared to conventional MRI, quantitative T1 and T2 maps generated by MRF showed additional findings in 4 of 15 patients, which were concordant with patients’ electroclinical presentation (Ma et al., 2019). While these studies highlight the promise of MRF, there is a lack of systematic reports on the radiological and clinical value of MRF as a supplementary imaging modality.
Methods:
MRF scans (1 mm
3 isotropic resolution) were acquired at 3T, generating 3D quantitative T1 and T2 maps for 71 patients who underwent presurgical evaluation at Cleveland Clinic Epilepsy Center between July 2017 and Nov 2023. MRF maps were reviewed in comparison to conventional 3T imaging by a consensus review between a board-certified neuroradiologist (SEJ) and an experienced neuroimaging specialist (ZIW). Findings were categorized as consistent, additional findings on MRF, or not recapitulated on MRF. We report these findings in relationship to the clinical, electroencephalography (EEG), surgical, and post-operative seizure outcome data.
Results: A total of 46 patients underwent surgery, 34 of whom became seizure-free (74%). MRF yielded additional findings in 28 of 71 patients (39%) and remained consistent with conventional clinical imaging in the remaining 43 patients (61%). In one patient, a left frontal finding seen on conventional MRI, was not recapitulated on MRF (1.4%). Additional benefits of MRF included: improved lesion conspicuity (N=10, 14% e.g. Figure 1); improved visualization of lesion extent (N=8, 11%) or mesial structure asymmetry (N=5, 7%, e.g. Figure 2), differentiation between multiple lesions (N=3, 4.2%), visualization of connection between lesion and other areas (N=2, 2.8%). Additional findings with MRF were more likely with certain pathology, including PVNH (3 patients with additional findings), and focal cortical dysplasia (specifically, FCD type IIa; 4 patients with additional findings).
Conclusions:
We report the first large cohort on the clinical utility of MRF as a supplement to conventional clinical imaging. MRF may provide additional benefit in visualizing the extent of cortical malformations (such as FCD), or differentiating between multi-lesional pathologies (such as PVNH). Findings from MRF may be important in guiding implantation strategies for invasive EEG or planning surgical resections.
Funding: NIH R01 NS109439