Additional file 1 of Oxygen-enhanced brain MRI using dynamic gradient-echo and MR fingerprinting

Additional file 1: Supplementary Table S1. The coefficients of variation of Pre-O2 and the slopes of the linear regressions for Pre-O2. Supplementary Table S2. Subject level of signal difference of (Post-O2 minus Pre-O2) of VIBE Supplementary Table S3. Signals of Pre-O2 and Post-O2 VIBE using VOIs eroded by 1 voxel from the original VOIs. Supplementary Table S4. T1 relaxation times of each dynamic phase using VOIs eroded by 1 voxel from the original VOIs. Supplementary Table S5. R1 relaxation times of each dynamic phase Supplementary Table S6. Ratios of T1 relaxation times relative to Pre-1 across dynamic phases. Supplementary Fig. S1 Representative images of 3D VOIs. CSF Cerebrospinal fluid, GM Gray matter, VOI Volume of interest, WM White matter. Supplementary Fig. S2 VOIs eroded by 1 voxel from the original VOIs (shown in Supplementary Fig. S1) were additionally created. Representative images of 3D VOIs. CSF Cerebrospinal fluid, GM Gray matter, VOI Volume of interest, WM White matter. Supplementary Fig. S3 Dynamic signal intensity changes on 3D VIBE for GM were presented with a thicker solid line and a narrower y-axis. A slight signal increase was also observed in GM. GM Gray matter, VIBE Volumetric Interpolated Breath-hold Examination. Supplementary Fig. S4 Slopes of Post-O2 VIBE (average of VIBE9th to VIBE20th) were shown. Slope of surface CSF was significantly highest among four VOIs. CSF Cerebrospinal fluid, VIBE Volumetric Interpolated Breath-hold Examination, VOI Volume of interest. Supplementary Fig. S5 Dynamic signal intensity changes on 3D VIBE using VOIs eroded by 1 voxel from the original VOIs. Temporal changes in the signal intensity ratio of VIBENst/VIBE2nd are shown across VOIs. A rapid increase in signal was observed in the surface CSF immediately after O2 inhalation. A slight signal increase was also observed in GM. Each circle represents the corresponding Participant 1-20. CSF Cerebrospinal fluid, GM Gray matter, VIBE Volumetric Interpolated Breath-hold Examination, VOI Volume of interest. Supplementary Fig. S6. Dynamic signal intensity changes on 3D VIBE using VOIs eroded by 1 voxel from the original VOIs for GM were presented with a thicker solid line and a narrower y-axis. A slight signal increase was also observed in GM. GM Gray matter, VIBE Volumetric Interpolated Breath-hold Examination, VOI Volume of interest. Supplementary Fig. S7 Cramér–Rao Lower Bound (CRLB)-based sensitivity analysis using the current sequence was performed to evaluate precision over a range of T1/T2 values, including those typical of CSF. In general, lower CRLB values indicate more precise parameter estimation. To provide a quantitative anchor for the long-T1 regime characteristic of surface CSF, the theoretical precision floor at T1 = 3,000 ms and T2 = 1,000 ms using a conservative target SNR of 10 was shown. Under these conditions, the lower bound for T1 estimation error, CRLB_T1, was 21.26 ms, corresponding to a relative CRLB of 0.71%. The observed Post minus Pre T1 reductions in surface CSF were approximately 150–250 ms, which are about 7–12 times larger than this CRLB-based precision floor. These results indicate that the observed T1 reductions substantially exceed the expected uncertainty from estimation noise and are unlikely to be explained by numerical precision limits alone.