Acute LTE electromagnetic field exposure modulates the human resting-state functional connectivity

Abstract

We designed a controllable near-field LTE EMF exposure environment (figure 1) [4]. 21 right-handed healthy subjects were enrolled to participate in a double, crossover, randomized and counterbalanced experiment including two sessions In each session, the EMF exposure (real or sham) lasted for 30 min. The structural and resting state fMRI signals were collected before and after the exposure. The head structural images were segmented interactively to identify 24 tissues with iSeg. Then we performed the Finite-Difference TimeDomain (FDTD) simulation [5] to estimate the specific absorption rate (SAR) power distribution induced in different head tissues. We examined the amplitude of low frequency fluctuation (ALFF) of brain activity [6], and found three brain regions showing significantly decreased ALFF values after real exposure [4]. Then we selected these three regions as seed volumes for the functional connectivity analyses. Data preprocessing included slice time correction, head motion correction, spatially normalization, linear detrend, Gaussian smoothing (FWHM = 4 mm), and temporal band-pass filter with 0.01 ~ 0.1 Hz. Then the mean time series of each seed volume were computed for each condition. Cross-correlation analysis was performed between the mean signals and the time series of all other brain voxels. After performing a Fisher's on correlation coefficients, we used paired t-test to generate the maps of statistical difference between pre-and post-exposure (P < 0.05, corrected with Monte Carlo simulation method). Figure 1. Experimental setup for LTE EMF exposure Figure 2. Slice views for estimated SAR power distribution in two subjects. SAR maps are colored according to the bar on the right, and 0 dB equals 2.36 W/kg for a specific voxel.