Research Article: Revealing heterogeneity of brain imaging phenotypes in Alzheimer’s disease based on unsupervised clustering of blood marker profiles

Date Published: March 4, 2019

Publisher: Public Library of Science

Author(s): Gerard Martí-Juan, Gerard Sanroma, Gemma Piella, Alfredo Vellido.


Alzheimer’s disease (AD) affects millions of people and is a major rising problem in health care worldwide. Recent research suggests that AD could have different subtypes, presenting differences in how the disease develops. Characterizing those subtypes could be key to deepen the understanding of this complex disease. In this paper, we used a multivariate, non-supervised clustering method over blood-based markers to find subgroups of patients defined by distinctive blood marker profiles. Our analysis on ADNI database identified 4 possible subgroups, each with a different blood profile. More importantly, we show that subgroups with different profiles have a different relationship between brain phenotypes detected in magnetic resonance imaging and disease condition.

Partial Text

Alzheimer’s disease (AD) is the most common cause of dementia, a condition affecting more than 47 million people worldwide [1]. AD is one of the biggest concerns in global health care, due to its large economic and social impact. It is characterized by a deposition of amyloid-beta (Aβ) protein in the brain and the formation of tau plaques [2], and its most prevalent symptom is a progressive decline and deterioration of cognitive skills, leading to death. AD has been characterized as a multi-factorial disease [2, 3], involving many different processes and biological phenomena. Despite many efforts spent on research, we know relatively little about many aspects of the disease.

We applied the proposed method to the described cohort of subjects from ADNI database, to find the subgroups with heterogeneous blood profiles and analyze the interactions with the disease of each subgroup using volume features and cortical thickness. All the experiments are reproducible following the instructions found in the repository of the project

We applied a multivariate data-driven procedure for AD subtyping using blood-based markers, to obtain heterogeneous groups with different blood marker profiles. We showed that patients with different profiles present different interactions between disease stage and brain phenotypes. Although existing blood markers can still not be used to properly diagnose the disease [7, 30, 31], using blood markers to detect patient profiles where the disease could behave differently and could provide valuable biological insights is a promising research direction.




Leave a Reply

Your email address will not be published.