Paolo Boffetta

University of Bologna

Full Professor

PLCO (Learn more about this study)

PLCO-1657

Sep 5, 2024

Using large-scale cohort studies to develop a novel AI tool for identifying lifetime, environmental and occupational determinants of healthy ageing (DARE project)

Non-communicable (or chronic) diseases represent the leading cause of morbidity, disability and mortality in high-income countries; in the elderly population, the main cause of overall mortality is represented by circulatory diseases, followed by cancer and respiratory diseases.
Our main hypothesis posits that employing a supervised clustering and unsupervised neural networks-driven methodology will allow us to unveil novel and previously uninvestigated dietary and lifestyle patterns, which significantly influence the incidence and progression of major aging-related health outcomes, specifically cancer, cardiovascular diseases, and cognitive decline. This hypothesis challenges traditional approaches that predominantly rely on predefined dietary and lifestyle patterns by employing a data-driven approach, aiming to uncover original risk profiles, which are expected to offer deeper insights into the complex interplay between nutrition, environmental, and lifestyle factors in the context of aging.
The project will be carried out by pooling and analyzing data from several cohorts at the international level. Data will be analyzed using unsupervised neural networks approach and supervised clustering for identifying novel patterns of risk factors associated with the following outcomes: cancer, cardiovascular disease and cognitive decline.