ALFF

ALFF, or amplitude of low-frequency fluctuations, is a metric used in resting-state functional magnetic resonance imaging (fMRI) to quantify spontaneous brain activity. It reflects the intensity of intrinsic low-frequency fluctuations in the BOLD signal, typically in the range of 0.01 to 0.08 Hz. ALFF is commonly applied voxel-wise to map regional spontaneous activity across the brain.

Calculation and workflow: After standard preprocessing (including slice timing, motion correction, normalization, and smoothing), the time series at each voxel is bandpass filtered to retain 0.01–0.08 Hz. The Fourier transform is computed to obtain the power spectrum, and the amplitude in the low-frequency band is derived as the square root of the power. ALFF is typically defined as the average amplitude within that band, effectively the mean fluctuation amplitude of the voxel’s time series in the low-frequency range. To improve comparability, ALFF maps are often standardized by dividing by the global mean ALFF or converted to z-scores (sALFF).

Variants: Fractional ALFF (fALFF) expresses the ratio of power within the low-frequency band to the total power across the detectable frequency range, reducing sensitivity to physiological noise. Different pipelines may also employ additional normalization steps to enhance cross-subject comparability.

Applications and interpretation: ALFF is used to identify brain regions with altered spontaneous activity in healthy individuals and in various conditions, including major depressive disorder, schizophrenia, Alzheimer’s disease, and autism. It can aid in characterizing state vs. trait differences and in mapping functional organization of the resting brain.

Limitations: ALFF is sensitive to preprocessing choices and physiological noise (e.g., respiration, heart rate). It is not a direct measure of neuronal activity and can be influenced by non-neuronal factors; results require careful interpretation and validation across datasets.