Use of computational techniques for the differentiation among healthy aging, mild cognitive impairment and Alzheimer's disease

• Grant number: 23/06563-9
• Support Opportunities: Regular Research Grants
• Start date: December 01, 2023
• End date: November 30, 2026
• Field of knowledge: Engineering - Biomedical Engineering
• Principal Investigator: Andriana Susana Lopes de Oliveira Campanharo
• Grantee: Andriana Susana Lopes de Oliveira Campanharo
• Host Institution: Instituto de Biociências (IBB). Universidade Estadual Paulista (UNESP). Campus de Botucatu. Botucatu , SP, Brazil
• City of the host institution: Botucatu
• Associated researchers: Angelo Quartarone ; Maria De Cola

Abstract

Alzheimer's disease (AD) can be understood as a degenerative and progressive dementia of the Central Nervous System, irreversible and of unknown cause. AD is mainly characterized by memory loss, disorientation in time and space, and accelerated intellectual deterioration. AD is the main dementia among the elderly over 65 years old and affects approximately 30 million individuals worldwide. AD begins in a stage known as Mild Cognitive Impairment (MCI), characterized, above all, by loss of recent memory and difficulty in reasoning. The progression of this condition can last for decades, from the appearance of its first signs to its most severe clinical symptoms, which include the loss of the ability to communicate, move and even eat. The clinical identification of the MCI makes its diagnosis difficult, since it is done through a battery of exhaustive, long-term cognitive tests, carried out under the supervision of a highly qualified professional, such as a neurophysiologist. In this sense, it is necessary to propose an automatic system for identifying MCI and monitoring its progression. Electroencephalography (EEG) is a non-invasive and low-cost technique capable of measuring the electrical potential arising from neuronal activities and, therefore, has been widely used in the investigation of AD progression. Furthermore, neuroimaging techniques, such as Magnetic Resonance Imaging (MRI) and Computed Tomography (CT), have played a key role in characterizing the clinical picture presented by patients with AD. Recently, a method based on the theory of complex networks showed high efficiency in identifying the advanced stage of AD using EEG signals. Quantile graph mapping, the name given to this method, is capable of robustly synthesizing the dynamic properties of time series in complex networks. In this work, the mapping in quantile graphs, proposed for use in signals (one-dimensional case), will be modified (extended) for its use in image analysis (two-dimensional case). Thus, MRI images in addition to EEG signals will be used to identify the first signs of AD. (AU)