PRECISION MEDICINE FOR RHEUMATOID ARTHRITIS: UTILIZING AIENHANCED MENDELIAN RANDOMIZATION IN BIOMARKER SELECTION

Objective: To investigate the role that precision medicine, biomarker selection, Mendelian randomization, and artificial intelligence have in the treatment of rheumatoid arthritis and to examine the correlation with clinical implementation and translation. Methods: The sample included 288 healthcare professionals from related clinical or biomedical fields, as well as researchers. Data were gathered on the following dimensions: Precision Medicine, Biomarker Selection, Mendelian Randomization, Artificial Intelligence, and Implementation/Translation, using a structured 30-item questionnaire. The answers were captured on a 5-point rating scale. The ShapiroWilk normality test, Cronbach's alpha, Kaiser–Meyer–Olkin test, Bartlett test, independent-samples t-test, one-way ANOVA, Kruskal–Wallis test, Chi-square test, Pearson correlation, and multiple linear regression were used in the statistical analysis. The significance level used for statistical tests was p<0.05. Results: Cronbach's alpha values obtained for the different scales of the questionnaire revealed high internal consistency ranging from 0.929 to 0.959, and an overall Cronbach's alpha value was 0.984. Sampling adequacy was excellent (KMO=0.983), while Bartlett's test was significant (χ²=9427.46, df=435, p<0.001). Professional role (p<0.001), age (p<0.001), and gender (p<0.001) were each shown to have significant differences. The correlation coefficients between all the main constructs of the study were positive and strong, ranging from about 0.819 to 0.852. Implementation/Translation was significantly and positively related to Artificial Intelligence (r=0.852, p<0.001). Precision Medicine, Biomarker Selection, Mendelian Randomization, and Artificial Intelligence were identified as significant positive predictors of Implementation/Translation, which explained 80.0% of the variance (R²=0.800, adjusted R²=0.797, p<0.001) in multiple regression analysis. Conclusion: Mendelian randomization combined with AI could be a useful approach to prioritizing biomarkers and precision medicines in rheumatoid arthritis. The principal constructs are strongly connected with each other and allow for further research on validated clinical and genomic data sets. These results are methodological and not clinical data, as the data used for the present analysis were synthetic/mock data.