No abstract

The availability of neuroimaging-based databases is helping immensely to understand the brain function in healthy and diseased conditions. This viewpoint highlights the objectives, commonalities, and differences within these existing databases and pointers for researchers to choose a particular database. We introduce a multimodal multidisease database, SWADESH, and its comparison with the existing databases. A futuristic database blueprint is proposed for housing multidisease, multimodal, and longitudinal brain imaging data systematically organized in a matrix form along with neuropsychological assessment scores for the identification of causal disease processes. The information-rich databases will ultimately assist with the systematic identification of prime features linked to causal disease processes, leading to the design of appropriate clinical trials for successful therapeutic interventions.

Alzheimer's disease (AD) is a devastating neurodegenerative disorder affecting millions of people worldwide. The etiology of AD is not known, and intense research involving multimodal neuroimaging data (e.g., MRI, functional MRI, PET etc.) is extensively used to identify the causal molecular process for AD. In this context, various imaging-based databases accessible to researchers globally, are useful for an independent analysis. Apart from MRI-based brain imaging data, the neurochemical data using magnetic resonance spectroscopy (MRS) provide early molecular processes before the structural or functional changes are manifested. The existing imaging-based databases in AD lack the integration of MRS modality and, thus, limits the availability of neurochemical information to the AD research community. This perspective is an initiative to bring attention to the development of the neuroimaging database, "ANSH," that includes brain glutathione (GSH), gamma aminobutyric acid (GABA) levels, and other neurochemicals along with MRI-based information for AD, mild cognitive impairment (MCI), and healthy subjects. ANSH is supported by a JAVA-based workflow environment and python providing a simple, dynamic, and distributed platform with data security. The platform consists of two-tiered architecture for data collection and management further supporting quality control, report generation for analyzed data, and data backup with a dedicated storage system. The ANSH database aims to present a single neuroimaging data platform incorporating diverse data types from healthy control and patient groups to provide better insights pertaining to disease progression. This data management platform provides flexible data sharing across users with continuous project monitoring. The development of ANSH platform will facilitate collaborative research and multi-site data sharing across the globe.

No abstract

Background: In vivo neuroimaging modalities such as magnetic resonance imaging (MRI), functional MRI (fMRI), magnetoencephalography (MEG), magnetic resonance spectroscopy (MRS), and quantitative susceptibility mapping (QSM) are useful techniques to understand brain anatomical structure, functional activity, source localization, neurochemical profiling, and tissue susceptibility respectively. Integrating unique and distinct information from these neuroimaging modalities will further help to enhance the understanding of complex neurological disease. Objective: To develop a processing scheme for multimodal data integration in seamless manner on healthy young population, thus establishing a generalized framework for various clinical conditions (e.g., Alzheimer’s disease). Methods: A multimodal data integration scheme has been developed to integrate the outcomes from multiple neuroimaging data (fMRI, MEG, MRS, and QSM) spatially. Furthermore, the entire scheme has been incorporated into a user-friendly toolbox- “PRATEEK”. Results: The proposed methodology and toolbox has been tested for viability among fourteen healthy young participants for bilateral occipital cortices as the region of interest. This scheme can also be extended to other anatomical regions of interest. Overlap percentage from each combination of two modalities (fMRI-MRS, MEG-MRS, fMRI-QSM, and fMRI-MEG) has been computed and also been qualitatively assessed for combinations of the three (MEG-MRS-QSM) and four (fMRI-MEG-MRS-QSM) modalities. Conclusion: This user-friendly toolbox minimizes the need of an expertise in handling different neuroimaging tools for processing and analyzing multimodal data. The proposed scheme will be beneficial for clinical studies where geometric information plays a crucial role in advance brain research.