Fig. 5
From: Machine learning prediction of future amyloid beta positivity in amyloid-negative individuals
Predicting future CSF-based and PET-based A\(\beta\)-positivity from baseline CSF and PET measures: a Bar plots showing the average AUC and average correlation score across 10 computation runs for predicting future CSF-based amyloid positivity, with 95% confidence intervals error bars. b Bar plots showing the average AUC and average correlation score across 10 computation runs for predicting future PET-based amyloid positivity, with 95% confidence intervals error bars. c Distribution of probability score derived by RLR for PET-based prediction in A\(\beta\)-Stable and A\(\beta\)-Converter groups with CSF and PET baseline measures. d, e Heatmap of coefficient values across 10 runs of 10-fold CV (100 models) for PET-based classification model (d) and pet-based regression model (e), with the bar graphs showing the importance of each predictor calculated by the mean of the absolute value of regression coefficient. The regression model is designed for predicting the difference between future and baseline global SUVR measures