Swadesh to host repository of big data on brain disorders

The facility named Swadesh will be operational soon and will have the basic capability to authenticate and process data of all people with neurological disorders.

The project proposed a big data architecture for the management of five modules: neurodegenerative diseases such as Alzheimer’s disease (AD), mild cognitive impairment (MCI) and Parkinson’s disease (PD); neuropsychiatrics such as schizophrenia and bipolar disorder; neurodevelopmental such as autism and epilepsy; Other disorders and healthy subjects.

“At present, the database has data of 500 AD and MCI patients and 70 PD patients. It also includes data of 600 healthy older persons and 800 healthy young persons in the control group,” said Professor Pravat Kumar Mandal, Director-in-Charge, NBRC.

According to researchers, many people may suffer from Parkinson’s disease post-Covid and many are already suffering from mental disorders.

Neuroimaging data of Covid-19 infected patients are also planned to be included in the database for analysis.

In the initial phase, Project Swadesh aims to integrate neuroimaging data generated through multimodal neuroimaging and neuropsychological assessment at NBRC and its partner institutions, All India Institute of Medical Sciences (AIIMS), New Delhi and Medanta Hospital, Gurugram.

“In the future, the project will be a repository of multidimensional neuroimaging data. To the best of our knowledge, this is the first large-scale multimodal neuroimaging database initiative to be specifically designed for different disease categories under a single platform with big data architecture and state-of-the-art data analytics. Designed for the Indian population, it is a sincere effort towards building and enriching a multidisciplinary neuroimaging platform for identification of specific biomarkers of various brain disorders,” Mandal said.

The researchers note that large amounts of neuroimaging data are being produced everyday around the world. It is geared towards “big data” because of its volume, complexity, value and diversity.

Data generated from various neuroimaging modalities such as magnetic resonance imaging (MRI), functional MRI (fMRI), magnetic resonance spectroscopy (MRS) and neuropsychological tests provide structural, functional, neurochemical and behavioral information, respectively.

Researchers around the world use this information to define a disease by analyzing its imaging-based biomarkers.

Neuroimaging datasets from around the world provide a vast amount of information for the study of various brain disorders, standardization of parameters for clinical settings, and in-depth case studies.

“Often, a one-sided approach is taken, whereby imaging data for a particular disease is compared with imaging data from a control group of healthy subjects to determine the differences exhibited by the diseased state,” Mandal said.

“However, a challenge is that many brain disorders can have overlapping symptoms, making it difficult to make a correct diagnosis with big data analytics. Furthermore, the available neuroimaging datasets are disease-specific,” he said.

“In order to derive meaningful insights from neuroimaging big data, it is imperative to organize structural, functional, neurochemical, genetic and behavioral data for different diseases under one roof. This (indigenous) will lead to effective clinical trials and will facilitate the development of medical science,” he said.

The research, led by Mandal and researchers from Neuroscience and Mental Health, Melbourne, Australia, aims to build on the National Initiative for Brain Research, Swadesh, to host multimodal neuroimaging data for various disease categories with data analytics tools.

Mandal said that with a team of people trained in a variety of engineering and psychological domains, the project benefits from a variety of perspectives.