Skip to main content
2025 SCMR Conference Main banner
Final Program


Schedule at a Glance
Back

Rapid Fire Abstracts

Rapid Fire 04- Multiparametric Mapping II

Test-Retest Reproducibility of Cardiac Magnetic Resonance Fingerprinting T1 and T2 Mapping at 3T: A Healthy Volunteer Study (RF_TH_279)

Thursday, January 30, 2025
5:50 PM – 6:00 PM East Coast USA Time
Location: Blue Room PreFunction
Background: Cardiac Magnetic Resonance Fingerprinting (cMRF) has emerged as a promising technique for simultaneous myocardial T1 and T2 mapping, offering several potential advantages over conventional approaches, such as enhanced efficiency and more precision(1, 2). Building on previous research that reports normal distributions cMRF-based simultaneous T1 and T2 mapping in healthy subjects at 1.5T,(3) We aimed to evaluate the performance of cMRF T1 and T2 mapping in healthy volunteers at 3T.


Methods: A prospective single-center study was performed in which healthy volunteers (N=39) were scanned twice on a 3T CIMA Siemens scanner using Modified Look-Locker Inversion Recovery (MOLLI) sequence for conventional T1 mapping, T2-prepared balanced steady-state free precession (bSSFP) for conventional T2 mapping, cMRF T1/T2mapping, and short axis CINE sequences utilized for feature tracking strain analysis. The cMRF acquisition used a prospectively ECG-gated (or peripheral pulse-gated), breath-held scan: voxel size of 1.6x1.6x8 mm3, spiral readout, and 254 ms acquisition window, and data were acquired over 15 heartbeats in three cardiac short axis slices. Conventional T1 and T2 maps were generated for basal, mid, apical short axis acquisitions, using inline reconstruction (Figure 1).  The cMRF data were reconstructed using iterative low rank reconstruction and scan-specific dictionaries that included slice profile and preparation pulse efficiency corrections. Test–retest reproducibility of cMRF and conventional T1/T2 mapping were assessed using intraclass correlation coefficient (ICC).  Bland Altman analysis was performed comparing cMRF vs conventional T1/T2 mapping.


Results:

We evaluated 39 healthy volunteers (15 males and 24 females). Female volunteers had smaller LV volume (71.2 VS 80.7 ml/m2, P-value 0.011), lower LV mass index (42.9 VS 49.0 g/m2, P-value 0.021), and less stroke volume index (39.9 VS 47.8 ml/m2, P-value 0.03). Female volunteers exhibited higher global T1 relaxation time in both cMRF (1427 VS 1395 ms, P-value 0.0071) and conventional technique (1241 VS 1211ms, P-value 0.0027) (Table 1).  ICC for conventional native T1 from cMRF were similar [0.81 (0.64-0.90) vs 0.75 (0.43-0.88)], with a trend toward higher ICC with cMRF. ICC for cMRF vs conventional T2 were similar [0.85 (0.70-0.92) VS 0.85 (0.71-0.92)]. Male volunteers showed a trend toward higher ICC values for both cMRF and conventional T1 [0.85 (0.51-0.95) VS 0.55 (-0.22-0.85)] but not T2 [0.96 (0.88-0.98) VS 0.95 (0.85-0.99)]. Bland Altman results demonstrated proportional bias of higher cMRF compared to conventional T1 (180ms ± 45ms, CI [93, 269ms]) and lower T2 times (-12 ± 3ms, CI [-18, -6ms]) (Figure 2).



Conclusion: cMRF T1 and T2 mapping demonstrate similar test/retest metrics compared to conventional T1/T2 mapping at 3T, with significant increased native T1 times in women vs men. cMRF measured at 3T demonstrate significantly increased T1 times and decreased T2 times compared to conventional T1/T2mapping, respectively.