Skip to main content
Fig. 5 | BMC Biology

Fig. 5

From: Deep learning-based behavioral profiling of rodent stroke recovery

Fig. 5

Kinematic changes in runway walk after stroke. A Schematic overview of analysis from the sides and time course of the experiment. B Stick profile of fore and hindlimb movement in individual mouse. C, D Walk profile of hindlimb and forelimb joints in intact and stroked mice; x-axis represents relative height and y-axis represents time across multiple steps (grey shades). E Absolute height of selected joints at baseline and 3, 7, 14, 21 dpi. F Experimental design of runway testing. G Protraction and retraction of joints throughout a time course. H Angular variability between front and hindlimb joints. I Random Forest classification of most important parameters. J Principal component analysis of baseline 3, 7, 14, 21 dpi. Data are shown as mean distributions where the white dot represents the mean. Boxplots indicate the 25 to 75% quartiles of the data. For boxplots: each dot in the plots represents one animal. Line graphs are plotted as mean ± sem. For line graphs: the dots represent the mean of the data. Significance of mean differences between the groups was assessed using repeated ANOVA with post hoc analysis. Asterisks indicate significance: *P < 0.05, **P < 0.01, ***P < 0.001. i-h-a, iliac crest-hip-ankle; h-a-t, hip-ankle-toe; s-e-w, shoulder-elbow-wrist; e-w-t, elbow-wrist-toe; PC, principal component

Back to article page