Abstract
In recent years, brain connectivity is gaining ever-increasing interest from the interdisciplinary research community. The study of brain connectivity is characterized by a multifaceted approach providing both structural and functional evidence of the relationship between cerebral regions at different scales. Although magnetic resonance (MR) is the most established imaging modality for investigating connectivity in vivo, the recent advent of hybrid positron emission tomography (PET)/MR scanners paved the way for more comprehensive investigation of brain organization and physiology. Due to the high sensitivity and biochemical specificity of radiotracers, combining MR with PET imaging may enrich our ability to investigate connectivity by introducing the concept of metabolic connectivity and cometomics and promoting new insights on the physiological and molecular bases underlying high-level neural organization. This review aims to describe and summarize the main methods of analysis of brain connectivity employed in MR imaging and nuclear medicine. Moreover, it will discuss practical aspects and state-of-the-art techniques for exploiting hybrid PET/MR imaging to investigate the relationship of physiological processes and brain connectivity.
Highlights
One of the main challenges in neuroscience is to gain an understanding of how brain activity generates behavior and, in more detail, how neural information is segregated and integrated
Anatomical and physiological studies support the idea that cognitive processes depend on interactions among distributed neuronal populations and brain regions (Sporns, 2013); brain connectivity (BC) refers to the structural and functional characterization of those interactions
The recent development of hybrid positron emission tomography (PET)/magnetic resonance (MR) scanners allows for more thorough investigation of BC and its underlying physiological processes (Wehrl et al, 2013; Riedl et al, 2014; Aiello et al, 2015; Tahmasian et al, 2015)
Summary
One of the main challenges in neuroscience is to gain an understanding of how brain activity generates behavior and, in more detail, how neural information is segregated and integrated. This method, thereafter referred to as Interregional Correlation Analysis (IRCA), is principally similar to SBA and estimates the correlation between mean values of glucose metabolic rate (GMR) of pre-defined brain regions This approach demonstrated MC networks for the first time and documented their potential as biomarkers in Alzheimer’s disease and many subsequent studies confirmed its suitability for MC research (Table 1). Sparse inverse covariance estimation (SICE; Huang et al, 2010; Zou et al, 2015) yields the correlation between a pair of ROIs, given all other regions, and further demonstrated the relevance of MC in AD studies Another global approach is the scaled subprofile model principal component analysis (SSM-PCA; Moeller et al, 1987) that was first successfully employed in AIDS dementia complex and widely employed for MC characterization of various diseases (Table 1). Intensity normalization (i.e., scaling of tracer uptake to a reference region) is in most cases essential for analyses of non-quantitative data, as is the case
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
