Wet lab → Web lab

Anisha Keshavan

University of Washington

eScience Institute & Institute for Neuroengineering

Child Mind Institute

The Brain.

Development

Mental Health

Pathology

Aging

We need a lot of data.

Because we are all so unique

Computers can store our data

source: flickr

We have algorithms

A markov chain (wikimedia)

Neuroinformatics: A new discipline?

PhD student

Neuroinformatics reality:

A set of existing (distinct) disciplines?

Maybe neuroinformatics should be a web of disciplines?

create a web, not another mountain

The web, quite literally

me, a UWIN postdoc

Pediatric Mental Health

Most mental health disorder are diagnosed at childhood and adolescence
How do brain tissue volumes change during development?
- are individual differences in development meaningful?

Healthy Brain Network

10,000 MRI scans

SO MUCH (statistical) POWER

a biobank from the Child Mind Institute

An MRI scan takes 5 min

Some kids can't stay still

Bad quality data

Bad inferences

Data needs visual inspection

10,000 MRI scans

Why not algorithms?

45 MPH

STOP!

source

we need some expertise

We should work together

Algorithms

fast
sometimes stupid
needs lots of data to learn

Humans

slow
sometimes smart
learns from sparse data

Braindr

Brain Data Review

braindr.us

Keshavan, Anisha, Jason Yeatman, and Ariel Rokem. "Combining citizen science and deep learning to amplify expertise in neuroimaging." bioRxiv (2018): 363382. (under review)