fig5

Figure 5. Biomarker identification using ELISA and ROC analysis in the validation cohort of patients with IS. The OE and LE classifications were based on 3-month Longshi Scale outcomes, where OE indicates patients who improved from Bedridden to Domestic or Community, and LE indicates those who remained Bedridden. (A) Distribution of age among the enrolled participants; (B and C) ADL assessment of the enrolled patients at admission and after 3 months by Longshi scale (1 = Bedridden group, 2 = Domestic group, 3 = Community group) and BI in the LE and OE groups. Each dot represents an individual; (D-F) Bar graphs showing serum protein levels of the three target molecules (MMP9, CEACAM1, and MCAM) among recruited participants in the LE (n = 30), OE (n = 22), and HCs (n = 20) groups. Data are presented as mean ± SD, with individual data points overlaid on the bars to illustrate the distribution of measurements; (G-I) Correlation scatter plots of target molecules vs. age. Each dot represents an individual. The data were analyzed using one-way ANOVA. ^*P < 0.05; ^**P < 0.01; ^***P < 0.001; and ^****P < 0.0001; (J-L) ROC analyses were performed as a binary classification using serum protein levels as continuous predictors. ROC curve analysis of MMP9, CEACAM1, and MCAM to distinguish between individuals in the LE and OE recovery groups. The black dots represent points on the ROC curve to gain optimal differentiation measures. BI: Barthel Index; LE: little-effect recovery group; OE: obvious-effect recovery group; HCs: healthy controls; IS: ischemic stroke; ROC: receiver operating characteristic curve; MMP9: matrix metalloproteinase 9; CEACAM1: carcinoembryonic antigen-related cell adhesion molecule 1; MCAM: melanoma cell adhesion molecule.