AgeML: Age modelling with Machine Learning

Abstract

AbstractA successful approach to age modeling involves the supervised prediction of age using machine learning from subject features. Used for exploring the relationship between healthy and pathological ageing in brain and multiple body systems, as well as the interactions between them, we lack a standard for prediction of age from any generic system. In this work we developed AgeML, an OpenSource software for age-prediction following wellestablished and tested methodologies from any type of tabular clinical data. The objective is to set standards for reproducibility and standardization of reporting in supervised age modelling tasks. AgeML allows for modelling age and calculating age deltas, the difference between predicted and chronological age, measuring correlations between age deltas and factors, visualising differences in age deltas of different clinical populations and classifying clinical populations based on age deltas. Using the software AgeML, we’re demoing its capabilities on a hybrid dataset, reproduce published work, and unveil novel relationships between body organs and polygenetic risk scores. AgeML made easy for standardization and reproducibility.