Geometric deep learning on graphs for electroencephalogram dataset

Geometric deep learning on graphs for electroencephalogram dataset

Geometric deep learning on graphs is a burgeoning field applied to analyze electroencephalogram (EEG) datasets, spearheaded by Bishesh Khanal and Harish Bhandari. Their collaborative efforts aim to revolutionize neurological disorder diagnosis by leveraging this innovative approach.

Problem:

  • Identification of neurological disorders presents significant challenges due to the reliance on costly diagnostic tools like MRI and fMRI.
  • Limited access to expert medical personnel further exacerbates this issue, especially in remote regions.¬†
  • Traditional methods fall short in providing affordable and accessible solutions for early detection and treatment.

Research Aim:

  • Develop and implement a tailored geometric deep learning model for EEG datasets.
  • Enhance diagnostic accuracy while reducing costs.
  • Improve accessibility to neurological disorder diagnosis, particularly in remote regions.

Outcome So Far:

  • Initial investigations show promise in applying geometric deep learning to EEG datasets.
  • Collaborative efforts have yielded encouraging results.
  • Potential for improved accuracy in diagnosis identified.
  • Ongoing research aims to refine and optimize the model for widespread implementation.

Team:

Bishesh Khanal and Harish Bhandari

Research Themes: Transforming Global health with AI (TOGAI)
Project Category: Machine Learning, Medical Imaging