IEEE Consumer Electronics Magazine - May/June 2021 - 9

Table 1. Example milestone timeline of biomedical research to commercialization.
(1)
(2)
Phases!
Applications #
DNA
Speech
Recognition
Brain Recognition
(3)
(4)
(4)-(3) vs. (4)-(1)
Discovery Prototype Data sharing Commercialization
Year
1869
1876
Retina Recognition 1935
1908
1953
1990
1981
1998
1980
1992
2001-2006
2004þ
1996
2017
2000-2020
2024#
Notes: 200 billion of iris pairs comparison; + ADNI 1 Initiation; # NIA target.
have realized that we need trustful and comprehensive
open large brain datasets with at least
10,000 subjects and a multiprone performance
metrics, including area under receiver operating
characteristic curve, equivalent error rate, and
accuracy to take advantage of CNN methodology.
As such, the current ADNI is insufficient for classification
under CNN, and the performance metrics
have yet to be standardized.
Historical biomedical research milestones
have proved that until massive data could be
cultivated for validation and reproduction, analogous
to having standard currency exchange
from a bank, any neurotechnology development
would not be positioned to directly benefit consumer
brain healthcare at large. Case in point is
illustrated in Table 1 for the time involved from
discovery through data bank establishment in
the making of a solid product. Table 1 shows
approximately 20 years from large-scale data
sharing to reach significant commercialization.
ADNI was originated in 2004. Could it be hopeful
that by 2024, treatments would be available for
AD patients? We hope that events similar to
BDBC can play an anchoring role toward such
realization for consumption by seven million AD
patients today in USA.
The BDBC-Finale had five teams delivering
presentations regarding AD. Topics presented
are as follows.
1. Utilizing deep learning model to predict brain
age for Alzheimer's disease and mild cognitive
impairment patients, by Team MINE Professor
Brain, National Central University,
Taoyuan City, Taiwan.
May/June 2021
2. Improve deep learning performance using
transfer learning to predict early stages of
Alzheimer in ADNI dataset, by Team Pokemon
Brain, University of Missouri, Kansas
City, MO, USA, which won the first place.
3. Identifying age and cognitive impairment
based on EEG data, by Team i-Pavlovian, Pavlov
Institute of Physiology, Russia Academy
of Sciences, Saint Petersburg, Russia, which
won the second place.
4. Functional connectivity of neural networks and
cognitive status in patients with Alzheimer's
Disease, by Team MTLNeuro, S.M. Kirov Military
Medical Academy, St Petersburg, Russia.
5. Neurophysiological effects on Alzheimer's
disease rehabilitation by Saffron and Curcuma
Longa using rats in Morris Water Maze
Experiments, by Team Visual Physiology,
National Academy of Sciences, Karayev Institute
of Physiology, Baku, Azerbaijan.
References and presentation slides including
the Keynote and the Panel discussion devoted to
user-centric design are accessible from https://
spcn2020taiwan.com/finale/.
N. Nan Chu is the co-Founder of CWLab International,
Thousand Oaks, CA, USA. She was the Consumer
Technology Society Representative on the
IEEE Brain Initiative, from 2016 to 2020. During
which time, she coordinated the Brain Data Bank
Challenge to hold 12 events around the world,
overseeing 39 international research teams scrutinizing
3 TB of MRI and 350 GB of EEG datasets.
Contact her at nchu.cwlab.com.
Years Required
16 versus 127
25 versus 140
14~19 versus 85
20 versus 116
9
https://www.spcn2020taiwan.com/finale/ https://www.spcn2020taiwan.com/finale/
IEEE Consumer Electronics Magazine - May/June 2021

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - May/June 2021
