IIT Guwahati Researchers Develop Novel Algorithm for Brain Network Analysis

The developed algorithm offers interpretability to MRI data, claiming to revolutionize the diagnosis and treatment of neurodegenerative disorders.

Researchers at the Indian Institute of Technology, Guwahati, have unveiled a novel algorithm named Unique Brain Network Identification Number (UBNIN), aimed at encoding the intricate brain networks of both healthy individuals and patients diagnosed with Parkinson's disease (PD).

The study, conducted in collaboration with the National Institute of Mental Health and Neurosciences (NIMHANS), India, involved the analysis of structural brain MRI scans from 180 PD patients and 70 healthy individuals.

By adopting a network perspective, the researchers represented various brain regions as nodes and established connection values based on regional grey matter volume.

These connection values were then weighted to capture the significance of each link through a series of algorithmic steps.

Further, the resulting numerical representation (UBNIN) was found to be distinct for each individual brain network and applicable to other neuroimaging brain modalities.

Potential Implications in Parkinson's Disease Management

Parkinson's disease, a neurodegenerative disorder characterized by symptoms such as tremors, stiffness, and slow movement, progresses with age.

However, neurodegeneration begins long before these symptoms manifest, highlighting the importance of early detection for effective management.

In a pioneering study, researchers from IIT Guwahati and NIMHANS utilized non-invasive structural MRI scans during rest to investigate the impact of age on brain connectivity across different age groups.

The findings revealed a decreasing trend in the clustering coefficient with increasing sparsity for each age cohort, shedding light on the evolving dynamics of brain connectivity in PD patients.

Explaining the research findings, Dr Cota Navin Gupta, assistant professor, Neural Engineering Lab, Department of Biosciences and Bioengineering, IIT Guwahati, said, "UBNIN is a special number representing unique characteristics of each human brain from a network perspective. Interestingly, we can also reverse engineer any human's UBNIN value to reconstruct the original brain network. This UBNIN algorithm will enable us to identify and characterize (encode-decode) brain networks of every human beings efficiently."

PhD Scholar Tanmayee Samantaray highlighted the potential of applying the UBNIN algorithm to longitudinal neuroimaging data to elucidate the dynamics of brain plasticity over time. This insight is essential for understanding how the human brain degenerates and copes with damage caused by neurological diseases.

The developed UBNIN algorithm offers interpretability to MRI data, claiming to revolutionize the diagnosis and treatment of neurodegenerative disorders potentially. It can serve as a biomarker, complementing other diagnostic tests neurologists recommend.

UBNIN's versatility extends from brainprinting to optimizing storage for structural MRI brain networks, paving the way for efficient information transfer in telemedicine and related fields.

Its adaptability also holds promise for applications in other neuroimaging modalities such as EEG and functional MRI, as well as in various neurological conditions including Schizophrenia, Alzheimer's, and Depression.

Additionally, its applicability to diverse datasets, including protein, social, and traffic networks, positions it as a versatile tool for understanding complex system dynamics.

Future Directions & Support

Dr Gupta highlighted ongoing efforts to explore the potential of UBNIN as a biomarker for distinguishing between healthy individuals and those with Parkinson's disease at a group level.

The research findings have been published in the journal Brain Sciences and co-authored by Tanmayee Samantaray, Utsav Gupta, Dr Jitender Saini, and Dr Cota Navin Gupta.

This research has received funding from the Ministry of Education (MoE) doctoral scholarship, Government of India. It is supported by the Scheme for Promotion of Academic and Research Collaboration (SPARC Grant), Government of India.