Automatic feature extraction with Vectorial Genetic Programming for Alzheimer's Disease prediction through handwriting analysis

Alzheimer's Disease (AD) is an incurable neurodegenerative disease that strongly impacts the lives of the people affected. Even if, to date, there is no cure for this disease, its early diagnosis helps to manage the course of the disease better with the treatments currently available. Even more importantly, an early diagnosis will also be necessary for the new treatments available in the future. Recently, machine learning (ML) based tools have demonstrated their effectiveness in recognizing people's handwriting in the early stages of AD. In most cases, they use features defined by using the domain knowledge provided by clinicians. In this paper, we present a novel approach based on vectorial genetic programming (VE_GP) to recognize the handwriting of AD patients. VE_GP is an enhanced version of GP that can manage time series directly. We applied VE_GP to data collected using an experimental protocol, which was defined to collect handwriting data to support the development of ML tools for the early diagnosis of AD based on handwriting analysis. The experimental results confirmed the effectiveness of the proposed approach in terms of classification performance, size, and simplicity.