Normalization of functional MRI signals

Although blood oxygenation level dependent (BOLD) fMRI provides a useful tool for probing neuronal activities, large inter-subject variations in signal amplitude are commonly observed. Understanding the physiologic basis for these variations will have a significant impact on many fMRI studies:

  • The physiologic modulator can be used as a regressor to reduce variations across subjects, thereby improving statistical power for detecting group differences. 
  • If a pathologic condition or a drug treatment is shown to change fMRI responses, monitoring this modulatory parameter can help interpret the fMRI changes to neuronal deficits/recruitments. 

We present evidence that the task-evoked fMRI signals are modulated by baseline blood oxygenation. To measure global blood oxygenation, we used a recently developed technique, T2-Relaxation-Under-Spin-Tagging MRI, yielding baseline oxygenation of 63.7±7.2% in sagittal sinus with an estimation error of 1.3%. It was found that individuals with higher baseline oxygenation tend to have a smaller fMRI signal and vice versa. For every 10% difference in baseline oxygenation across subjects, the BOLD and cerebral blood flow signal differ by -0.4% and -30.0%, respectively, when using visual stimulation. TRUST MRI is a useful measurement for fMRI studies to control for the modulatory effects of baseline oxygenation that are unique to each subject.

Next, we aim to test the utility of venous oxygenation normalization in distinguishing subject groups. A “model” condition was used in which two visual stimuli with different flashing frequencies were used to stimulate two subject groups, respectively, thereby simulating the situation of control and patient groups. It was found that visual-evoked BOLD signal is significantly correlated with baseline venous T2 (p=0.0003) and inclusion of physiologic modulator in the regression analysis can substantially reduce p values of group-level statistical tests. When applied to voxel-wise analysis, the normalization process can allow the detection of more significant voxels. The utility of other basal parameters, including blood pressure, heart rate, arterial oxygenation and end-tidal CO2, in BOLD normalization was also assessed and it was found that the improvement was less significant. Time-to-peak of the BOLD responses was also studied and it was found that subjects with higher basal venous oxygenation tend to have slower BOLD responses.

Brain scans
Group-level activation maps with un-normalized and normalized fMRI signals.
(a-c) Activation maps comparing 8Hz flashing checkerboard and fixation (n=10). Threshold: p value < 0.005, cluster > 800 mm3.
(c-d) Difference maps comparing two subject groups receiving 8Hz (n=10) and 4Hz (n=10) stimulus, respectively. Threshold: p value < 0.05, cluster > 800 mm3.
Results are presented from two normalization approaches, one regression-based and one model-based. Only activations in thalamus and occipital lobes are shown.