Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

BRAIN AS A COMPLEX SYSTEM

No description
by

Edu Sesma Caselles

on 21 April 2016

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of BRAIN AS A COMPLEX SYSTEM

Summary
...is our brain changing every time?
ENGINEERING APPROACH
LINEAR:
Ax1 + Bx2

TIME-INVARIANT:
x(t)
x(t-1)
PERCEPTION
BRAIN AS A COMPLEX SYSTEM
Graph theory
neuron imaging
Watts and Strogatz (1998) introduced their small-world network model
provided regional specialization
with efficient global information transfer

Psychology:
break down into elements
explain behavior in terms of simple causal relationships
group attempted to explain behavior as a function of groups
factors constituting a dynamic whole.
Aim: Investigating Communitarian
dynamic nature of
a complex system
cannot be understood
by thinking of the system
as comprised of
independent elements
Can a system like the brain be understood
via knowledge about individual neurons
A well-connected system comprised of various regions
that interact with each other to produce complex behaviors
Network Construction
What are considered nodes in the network
What are considered links or connections in the network
Are strong and weak links differentiated
Is there directedness in the network
BRAIN NETWORK REVIEW
Network nodes and links
Each neuron as a node with every synapse functioning as a link
Not possible to image or record from the
estimated 86.000.000.000 neurons
In average 7000
synapses
Anatomical Labeling atlas
physical connections that exist between brain regions (brain tissues as nodes)
underlying structural connectivity
alone does not determine functional connectivity
Functional Network
assess the consistent functional changes between different regions
What we are gonna talk about from now on!
Functional network
Weighted links:
size,density, coherence of anatomical tracts in anatomical networks
Negative links ?
links:
strength of correlation or
causal interactions in functional networks
Unweighted links:
applying a
threshold
and discretizing on a weighted network
Threshold
No threshold:
Fully connected network

important role in network construction
important when comparing networks
fixed threshold
fixed average degree
fixed edge density
5% significance level (omit spurious)
absolute threshold level across networks
keeps all nodes connected to the main component of the network



connectivity have different average degree
across individuals or conditions
same correlation coefficient cut-off across networks

different absolute threshold

networks with generally weaker connectivity values
include more weak links

networks with higher connectivity values may omit strong links

The proportion of the number of actual edges to the total possible number of edges


Similar to fixing average degree
show how graph metrics change over various thresholds and select a threshold based on the size of the network


size-density relationship among self-organized networks that follows a power law
difficult challenge!
Solutions:
fully weighted,no threshold but too much computational power
Information Flow in a Network
replication (copying information)
transfer (moving information)
serial (one node at a time)
parallel (simultaneously duplication at several nodes)
brain likely uses
parallel duplication
to transmit signals (30% of all synapses in a stochastic manner)
Graph Centrality
Classical node centrality:
degree centrality
betweenness centrality
closeness centrality
eigenvector centrality
it is important to question what is being measured
suggestions:
Centrality measures vary along four different
dimensions
choice of summary measure
type of walk considered
property of walk assessed
type of involvement


Challenge: proper choice of a centrality measure is heavily dependent on how information flows through the network.
Variety: 30 centrality metrics
Two major category:
Radial measures
Medial measures
assess movement of information that emanates from, or terminates at a given node
assess the number of walks that pass through a given node
high-degree centrality
localize to different areas of the brain (hubs)
implicated in various disease states, such as Alzheimer’s

Community Structure
groups of nodes that have more interconnections with each other than other nodes
help segregate the system into smaller compartments
Why Community:
groups can exhibit similar global properties, yet show substantial differences in community organization
In systems such as the brain, the actual community structure is not known
Modularity, Q

Order Statistics Local Optimization Method (OSLOM)

Modularity, Q is the most popular one (year 2012).
The adult human male brain has:

- 86 billion neurons

- That results in 125 trillion synapses
(at least 1,000 times the number of stars in our galaxy) with 1,000 molecular-scale switches each one.

- 125,000 trillion switches in a single human brain.
Modularity:
To determine the optimal hierarchical level
OSLOM
finds communities in a network based on statistical significance.
Node may exist in more than one community
Methods for community detection:
4,000,000,000 years to map it
"We will be able to model the entire human brain within 10 years..."
Henry Markram (2009)
HUMAN BRAIN'S PROJECT
“As I understand it, tons of data will be put into a supercomputer and this will somehow lead to a global understanding of how the brain works. ... To simulate the brain, or a part of the brain, one has to start with some hypothesis about how it works.” - Edvard Moser
it could be the entire computational capacity of our planet right now!!
"Our brain is the most complex machine that ever existed..."
...that founding should be use in a better research approach.
HUMAN BRAIN'S PROJECT
But anyways, it could be possible in a future...
ALGORITHM / SOFTWARE
HARDWARE?
OK... the scalability is huge,
but...
why complex?
Ay1 + By2
y(t)
y(t-1)
- GENETIC and PHYSICAL ASPECTS

- EXPERIENCES and LEARNING
NO LTI
COMPLEX
Previous thought:
"...neurons cannot reproduce after the first few years of life."


New research:
Neurogenesis discovered in the brain of 72-years-old adult!


Guide the process of neuronal growth to repair areas of the brain:

NEUROPLASTICITY
HOW TO GUIDE IT?
dopamine = pleasure
REWARD
the brain's reward pathway encourages us to seek out activities essential to species survival...
the reward pathway is activated, the brain floods with dopamine... we feel good, we seek to repeat the activity.
Rewiring the brain...
"neuro" = brains, "plastic" = changeable
Passion:
- just wait for dopamine!

Procedure:
- practice, practice and practice
TWO WAYS:
At the end, no matter how you rewired your brain...


...repetition and practice make these pathways stronger inside the brain.
Summary
SYNESTHESIA
...are we "in time"?
only allow a node to belong
to one community
Investigation of the dynamic changes
measures how stable a node is in a community over time or multiple realizations

how the modularity changes over time for a learning task

determine network community consistency across multiple realizations of the same network
Network Comparison
Focusing solely on the
degree of each node
fails to acknowledge the regions to which it is connected.
Highlights:
Rationale for using network science in the brain is to understand the organization and dynamics of the brain as a complex system

Measurements within a network are not independent

No nodal measurement in a network is completely independent (coefficient,etc..)

Comparing summary metrics across the brain may not provide suitable information about a group
Presentation guideline
Using Network Science as a Tool for Understanding the Brain
Interesting Properties Of The Brain
References:
Telesford, Qawi K., Sean L. Simpson, Jonathan H. Burdette, Satoru Hayasaka, and Paul J. Laurienti. "The brain as a complex system: using network science as a tool for understanding the brain." Brain Connectivity 1, no. 4 (2011): 295-308.
Newman MEJ. 2006. Finding community structure in networks
using the eigenvectors of matrices. Phys Rev E Stat Nonlin
Soft Matter Phys 74:036104.
Petrella JR. 2011. Use of graph theory to evaluate brain networks:
a clinical tool for a small world? Radiology 259:
317–320.
A well-connected system comprised of various regions
that interact with each other to produce complex behaviors
fixed threshold

fixed average degree

fixed edge density
..."how do you think" when you think about chocolate?
BRAINPRINTS
real time vs. memory
...how fast can you process it?
...how much energy your brain has to burn?
sometimes, time goes soooo slow...

...but we don't remember it
... this cherry tastes pretty "blue" !!
- Perceptual genome

- All the people have some level...
...but not all of them know it
(1948)
Concept: classify nodes as central, or more important, within a system.
(Freeman, 1977)
(Freeman, 1977)
(Bonacich, 1987)
(Freeman, 1977)
(Borgatti and Everett 2006)
(Achard et al., 2006)
(He et al., 2008; Buckner et al., 2009)
performed worst in identifying and differentiating hubs in the brain
( Joyce et al., 2010)
(Fortunato, 2010; Girvan and Newman,2002)
(Moussa et al., 2011).
(Lancichinetti et al., 2011).
(Palla et al., 2005),
to evaluate the consistency of network structure.
(Bassett et al., 2011)
Steen and associates (2011)
remains a fertile area for methodological development
Laurienti and associates (2011)
(Achard et al., 2006; Stam, 2004; van den
Heuvel et al., 2008; van Wijk et al., 2010)
...are we "sensing" enough?
20 Hz - 20 KHz
VEST
Versatile Extra-Sensory Transducer
THANKS FOR LISTENING!
Rationale for using network science in the brain is to understand the organization and dynamics of the brain as a complex system
Not only complex, also HUGE.
It changes: non-static system.
It depends of the inputs.
Different interpretations.
In two separate studies on epilepsy, one group reported lower path length between patients with epilepsy and the control group ; whereas another study showed an increase
(Liao et al., 2010)
(Vlooswijk et al., 2011)
Network Comparison
is lower path length a sign of depression or multiple sclerosis
in a study comparing the effects of an exercise regimen in a group of older adults, no significant differences were found for clustering or path length, but the topology of the networks was dramatically different
(Burdette et al., 2010)
Brain cannot be summarized by a single number
threshold

1986-2012 : "
Connectome
"
...a map of all 302 neurons in the C. Elegans nervous system as well as all the 7,000 connections.
QUESTIONS
Can Deep Neural Networks reveal the complexity of Neural Representations across the Brain's community Pathways?


If we could map the human brain in a future...
Is mapping the human brain worth it? How far is the idea of understanding the brain from just mapping it? Will our technology be able to handle the brain behavior?


In a similar way a deaf person can hear through the (versatile extra-sensory transducer) vest, will a blind person ever be able to see? Also, can you think of any other additional senses that could be useful?
Full transcript