We used deconvolution analysis to examine temporal changes in brain activity after acupuncture stimulation and assess brain responses without expected reference functions. We also examined temporal changes in brain activity after sham acupuncture (noninsertive) and scrubbing stimulation. We divided 26 healthy right-handed adults into a group of 13 who received real acupuncture with manual manipulation and a group of 13 who received both tactical stimulations. Functional magnetic resonance imaging (fMRI) sequences consisted of four 15-s stimulation blocks (ON) interspersed between one 30-s and four 45-s rest blocks (OFF) for a total scanning time of 270 s. We analyzed data by using Statistical Parametric Mapping 8 (SPM8), MarsBaR, and Analysis of Functional NeuroImages (AFNI) software. For statistical analysis, we used 3dDeconvolve, part of the AFNI package, to extract the impulse response functions (IRFs) of the fMRI signals on a voxel-wise basis, and we tested the time courses of the extracted IRFs for the stimulations. We found stimulus-specific impulse responses of blood oxygen level-dependent (BOLD) signals in various brain regions. We observed significantly delayed and long-sustained increases of BOLD signals in several brain regions following real acupuncture compared to sham acupuncture and palm scrubbing, which we attribute to peripheral nocireceptors, flare responses, and processing of the central nervous system. Acupuncture stimulation induced continued activity that was stronger than activity after the other stimulations.
We used tent function deconvolution to process fMRI data for acupuncture stimulation and found delayed increasing and delayed decreasing changes in BOLD signal in the somatosensory areas and areas related to pain perception. Deconvolution analyses with tent functions are expected to be useful in extracting complicated and associated brain activity that is delayed and sustained for a long period after various stimulations.