The IEEE Brain Data Bank Challenge 2020

Abstract

The 2020 Brain Data Bank (BDB) Challenge reached its closure on December 5, 2020, in a virtual conference format, due to COVID-19. Having 88 registrants (roughly 20% medical, non-IEEE, and the rest primarily IEEE engineering professionals) from 10 countries, the BDBC-Finale program consisted of presentations from 5 selected teams resulting from 3 preliminary runs earlier in 2020. The program was further enriched with a National Institute of Aging (NIA) Keynote on US funding sources and a Special Panel on Codesigning with an End-User, Prof. Justin Yerbury, who has been suffering from motor neuron disease since 2016. The theme for 2020 was the Aging Brain, making use of large brain records, (1) Alzheimer’'s Disease NeuroImaging (ADNI) containing magnetic resonance imaging (MRI), positron emission tomography, CerebroSpinal Fluid images, and clinical data (~3 TB), and (2) ElectroEncephaloGraphy (EEG) corpus (48–350 GB) to examine brain degeneration ending in Alzheimer's Disease (AD). Individual teams specified selected brain datasets to apply deep learning/machine learning techniques for the detection of elderly brain impairment. Other approaches to understand AD were also presented in terms of fMRI-based functional connectivity alteration, and experiments in rehabilitation using turmeric administered rats after bulbectomy. The author discuses the gains from these BDB Challenge presentations.