The vast potential and bright future of neuroimaging

Abstract

Significant advances in anatomical and functional neuroimaging techniques have allowed researchers and clinicians to visualize the brain in action. The field of neuroimaging currently includes newer and faster scanners, improved image quality, higher spatial and temporal resolution and diverse methods of acquisition and analysis. Beyond simply imaging brain structures, these developments enable quantitative assessment of the microstructural and functional architecture, perfusion and metabolism of the brain. The resultant highly granular data have the potential to greatly improve characterization of neurological, neurosurgical and psychiatric disorders without invasive neurosurgery. However, the surge in neuroimaging data that can be collected over a relatively short acquisition period has led to a "big data" problem, where novel methods are needed to appropriately extract and analyze information and integrate data with other types of big data, such as genomic and proteomic data. Another challenge is the translation of these new technologies from basic science into clinical practice, so that they can be leveraged to improve patient outcomes and alleviate human disease. Critical to this endeavor is research comparing the effectiveness and outcomes of these advancements to allow widespread acceptance in the modern, economically constrained healthcare system. This review aims to illustrate the different facets of cutting edge neuroimaging techniques, as well as the potential role of these methods as clinical tools for evaluating the breadth of diseases that affect the brain.

Figure 1.

PET-MRI scan obtained in a 73-year-old male patient, APOE-E4 homozygous, 90–110 min after administration of 5.4 mCi Florbetaben (Neuraceq). While this patient was cognitively normal, the positive beta amyloid scan (lack of differential uptake between cortical gray matter and white matter) together with his APO-E4 status, puts him at increased risk to develop Alzheimer’s disease. PET, positron emission tomography.

Figure 2.

Imaging genomics for brain tumors consists of combining imaging phenotype with genotype in order to improve the characterization of these tumors as well as outcome prediction in patients affected by these tumors.

Figure 3.

Big data approach to neuroimaging data allowing to extract quantitative neuroimaging features. dw, diffusion-weighted; MEG, magnetoencephalogram; BOLD, blood oxygen level-dependent.