Pimsoft 1

Laser Doppler measurements were made at a sample rate of 33 samples per second, and averaged over 1-min intervals using Perisoft for Windows, version 2.5.5 (Perimed AB, J€ arf€ alla, Sweden). The mean perfusion and the RBC concentration were measured with LSCI and PLSI, respectively, in five regions of both the irradiated and the non-irradiated breast: the upper inner quadrant (UIQ), lower inner quadrant (LIQ), upper outer quadrant (UOQ), lower outer quadrant (LOQ), and areola (A) (Figure 1). The mean perfusion in these regions was calculated by selecting regions of interest in the perfusion images using the LSCI system's software (PimSoft 1.3, Perimed AB, J€ arfalla, Sweden). Mean PLSI indices (RBC concentrations) were calculated from PLSI image files in the same manner as above using ImageJ (version 1.49t, US National Institutes of Health, Bethesda, MD). The inner quadrants and areola were measured from the frontal images, while the outer quadrants were measured from the lateral images.

All data in the text are given as mean (SD). Images were analyzed using the LSCI system's software (PimSoft 1.5, Perimed AB, Järfalla, Sweden). Regions of interest (ROI) were selected manually in the first image of each series. Then, the locations of ROI in subsequent images were verified and corrections were made as needed. The day-to-day and the site-tosite variability was analyzed using correlation analysis and with Bland-Altman plots. Normal distribution of the data was tested using D'Agostino & Pearson omnibus normality test. As the perfusion data obtained with different concentrations of MN was not normally distributed, differences between concentrations were analyzed with a Dunn's multiple comparisons test.
The data from different anatomical sites were normally distributed and were analyzed with a one-way ANOVA with multiple comparisons using the Tukey correction method. A twotailed paired Student's t-test was used to compare the difference in perfusion between the control and MN area. Statistical calculations were performed using GraphPad Prism version 6 for Windows (Graphpad Software, San Diego, CA, USA). A probability of less than 0.05 was accepted as significant.

Images were analyzed using Pimsoft software (PimSoft 1.5, Perimed AB, Järfalla, Sweden). The selection of ROIs was based on relative position to bregma to assure reproducibility between subjects. CBF data from regions of interest (ROI) was then exported to spike2 software for quantification. MAP data was analyzed with spike2 software (Cambridge Electronic Design, Cambridge, UK,version 6.15). CBF and MAP changes were measured for each stimulus intensity within a 30s window post-stimulus.
Responses amplitude was quantified as the peak value relative to the mean signal value of the 30 s artifact-free baseline prior to stimulus onset.