An Information Visualization Approach to Classification and Assessment of Diabetes Risk in Primary Care

摘要: Chronic disease risk assessment is a common information processing task performed by primary care physicians with many at-risk patients. However, effectively integrating about factors across patients cognitively difficult. Methods for visualizing multidimensional data may augment clinical providing reduceddimensional displays which stratify patient points according to level while additional insight into clinically important individual factor variables. This study combines medical evidence, dimensionality reduction techniques and visualization develop new framework visually classifying interpreting data. then explored analytically validated using unique health database from the American Diabetes Association that contains predictions made Archimedes model. Results show generate models classify large population accuracy comparable statistical methods. Further, visualizations provide rich give (i) relative importance of factors, (ii) confidence in (iii) overall distributions population. The proposed approach produce can be embedded systems interactive visual analysis tools support physician decision making.