Skip to main content
. 2021 Jul 20;10(7):1829. doi: 10.3390/cells10071829

Figure 3.

Figure 3

Results from the VBM multivariate linear regression analyses performed in CAT12 for the three groups (HC, MCI, AD). The results are covaried for total intracranial volume (TIV), age and education. The scatterplot displays the relationship between Locus Coeruleus (LC) volume (y axis) and TMT-A time in seconds (x axis) for the three groups: green (HC, n.395), orange (MCI, n.156), red (AD, n.135). On the x axis (TMT-A) are the seconds required to complete TMT-A. More seconds spent in completing TMT-A mirror a slower visuo-spatial cognitive processing related to the LC decline. The systematic decline of the LC volume across the three groups is related to a slower visuo-motor attentional performance. On the left portion of the figure, blue shows the average LC results for the three groups (n = 686, p < 0.001) on a 3D fronto-lateral view of the Brainstem and the Diencephalon. On the right portion of the figure are the 3D reconstructions (displayed in the MNI152 space) of the results in comparison with previously published LC atlases and masks. (a) average LC result; (b) average LC result is shown in the LC “omni-comprehensive” mask; (c) Keren et al. (2009) [102,107]; (d) Tona et al. (2017) [108]; (e) Betts et al. (2017) [109]; (f) Dahl et al. (2019) [33]; (g) Liu et al. (2019) [110]; (h) Rong Ye et al. (2020) [111]; (i) LC meta-mask by Dahl et al. (2021) [70]. The last column on the right shows the regions of the LC mask negatively related to TMT-A performances for the three groups considered separately (p < 0.01): HC n = 395 (no results), MCI n = 156 and AD n = 135.