Physiological noise in oxygenation‐sensitive magnetic resonance imaging

Abstract

The physiological noise in the resting brain, which arises from fluctuations in metabolic-linked brain physiology and subtle brain pulsations, was investigated in six healthy volunteers using oxygenation-sensitive dual-echo spiral MRI at 3.0 T. In contrast to the system and thermal noise, the physiological noise demonstrates a signal strength dependency and, unique to the metabolic-linked noise, an echo-time dependency. Variations of the MR signal strength by changing the flip angle and echo time allowed separation of the different noise components and revealed that the physiological noise at 3.0 T (1) exceeds other noise sources and (2) is significantly greater in cortical gray matter than in white matter regions. The SNR in oxygenation-sensitive MRI is predicted to saturate at higher fields, suggesting that noise measurements of the resting brain at 3.0 T and higher may provide a sensitive probe of functional information. Several biophysical models (1,2) and recent investigations of the field dependency in MRI (3-8) strongly suggest improvements in the signal-to-noise ratio (SNR) and the contrast-to-noise ratio (CNR) at higher magnetic fields. In a recent work (8), however, we have shown that the physiological noise demonstrates an MR signal strength dependency and exceeds the thermal noise and scanner noise at 3.0 T. Consequently, the physiological noise counteracts the gain in SNR at higher fields. In the present study, a model for the intrinsic noise in oxygenation-sensitive MRI of the human brain is developed. In corresponding experiments the MR-signal strength was modulated by variations in the flip angle (␣) and echo-time (TE) to separate individual noise contributions in the resting brain. Results were compared with the predicted noise contributions and corresponding implications on neuroimaging modalities are discussed. THEORYThe signal and the intrinsic noise in high-field MRI have been shown to be quadratic and linear in B 0 , respectively, resulting in an SNR proportional to the strength of the magnetic field (3,4). Whereas these earlier studies characterize the dominant noise by thermally generated random noise from the subject and scanner electronics and assume that coil losses are negligible (4), we expand this model by considering physiological noise as a further significant contribution to the total image noise:Here, T is the thermal noise from the subject and scanner electronics and S describes other system noise that includes drift and imperfections in RF, gradient, and shim subsystems. 0 , the square-law sum of T and S , is considered the raw noise and has been shown to be proportional to B 0 (4) but independent of the MR-signal strength. The term P in Eq.[1] describes the physiological noise, which arises from fluctuations in the basal cerebral metabolism (CMRO 2 ), blood flow (CBF), and blood volume (CBV), but also from cardiac and respiratory functions that cause quasiperiodic oscillations in the vascular system (9 -11), motion from subtle brain pulsatility (12), and magnetic field modula...

A component based noise correction method (CompCor) for BOLD and perfusion based fMRI

“…When utilized together, the BOLD and perfusion signals can provide a quantitative understanding of the metabolic response to neural activity and provide insight into neurovascular coupling mechanisms (Hoge et al 1999). However, as the fMRI community has moved to higher field strengths, physiological noise has become an increasingly important confound limiting the sensitivity and the application of fMRI studies (Kruger and Glover 2001;Liu et al 2006). …”

Influence of physiological noise on accelerated 2D and 3D resting state functional MRI data at 7 T

“…Because segmented 3D EPI approaches are more susceptible to physiological fluctuations than 2D sequences, the advantage of using 3D accelerated sequences may not directly translate into a higher image SNR. This sensitivity to noise should decrease as the temporal resolution increases (14,30) and is expected to become negligible for highly accelerated sequences. For SMS-EPI, several algorithms (8,(31)(32)(33)(34)(35) have been proposed to reconstruct the accelerated data.…”

Generalised filtering and stochastic DCM for fMRI

“…In other words, the only uncertainty arises at the point of observation, through measurement noise. However, many studies suggest that physiological noise due to stochastic fluctuations in neuronal and vascular responses need to be taken into account (Biswal et al, 1995;Krüger and Glover, 2001;Riera et al, 2004). Recently, there has been a corresponding interest in estimating both the parameters and hidden states of DCMs based upon differential equations that include state-noise.…”