Skip to main content
. 2023 Mar 15;38:103377. doi: 10.1016/j.nicl.2023.103377

Fig. 4.

Fig. 4

Neurovascular coupling with linear least-squares deconvolution. (a) Hemodynamic response function (HRF) before (top) and 2 days after stroke (bottom) in the forelimb and stroke regions outlined in green and red respectively. (b) Time course of 4 stimulation trials showing measured GCaMP signal overlaid with measured HbT and predicted HbT, obtained by convolving the GCaMP signal with the HRF kernel, at pre-stroke and day 2 for the regions outlined in (a). (c) Pearson’s correlation coefficient for measured HbT and predicted HbT for pre-stroke (top) and 2 days after stroke (bottom). (d) Regions used to extract HRF in (e) and (f). (e) HRF obtained by deconvolution model for HbT, HbO, and HbR, for one example mouse at each time point before and after stroke. Note the deviation in HRF compared to pre-stroke in the acute phase within the stroke and peri-infarct, and a return to pre-stroke HRF at week 4. (f) Same as in (e) for all mice (n = 12). Each line represents the HRF for one mouse. (g) Pixel-by-pixel Pearson’s correlation coefficient between measured and predicted HbT (top), HbO (middle), and HbR (bottom). Predicted HbX is obtained by convolving the GCaMP signal at each time point with a mean HRF obtained from pre-stroke data. (h) Pearson’s correlation coefficient quantified across all mice within the stroke core, peri-infarct, and contralesional forelimb region. A two-sample t-test was run for all statistical tests. Thick bars: p < 0.01, thin bars: p < 0.05. Note the sustained reduction of correlation coefficient within the stroke core but recovery within the peri-infarct for HbT and HbO. Yellow asterisk indicates stroke hemisphere. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)