Machine Learning and Social Network Analysis Applied to Alzheimer's Disease Biomarkers

Javier      Di Deco; Ana      M. Gonzalez; Julia      Diaz; Virginia      Mato; Daniel      Garcia–Frank; Juan      Alvarez–Linera; Ana      Frank; Juan      A. Hernandez–Tamames

doi:10.2174/1568026611313050008

Abstract

Due to the fact that the number of deaths due Alzheimer is increasing, the scientists have a strong interest in early stage diagnostic of this disease. Alzheimer's patients show different kind of brain alterations, such as morphological, biochemical, functional, etc. Currently, using magnetic resonance imaging techniques is possible to obtain a huge amount of biomarkers; being difficult to appraise which of them can explain more properly how the pathology evolves instead of the normal ageing.

Machine Learning methods facilitate an efficient analysis of complex data and can be used to discover which biomarkers are more informative. Moreover, automatic models can learn from historical data to suggest the diagnostic of new patients. Social Network Analysis (SNA) views social relationships in terms of network theory consisting of nodes and connections. The resulting graph-based structures are often very complex; there can be many kinds of connections between the nodes. SNA has emerged as a key technique in modern sociology. It has also gained a significant following in medicine, anthropology, biology, information science, etc., and has become a popular topic of speculation and study.

This paper presents a review of machine learning and SNA techniques and then, a new approach to analyze the magnetic resonance imaging biomarkers with these techniques, obtaining relevant relationships that can explain the different phenotypes in dementia, in particular, different stages of Alzheimer's disease.

Keywords: Alzheimer's Disease, Feature selection, Machine learning, Magnetic resonance imaging, Social network analysis.

« Previous Next »