See Also: Book Notes, (me), Notes on Consciousness

Notes on Resting State 2014

Fourth Biennial Conference on Resting State / Brain Connectivity, MIT, 2014.

restingstate.com <> 2014 Program <> wikipedia: Resting State fMRI

I attended the Resting State / Brain Connectivity last Thursday, Friday and Saturday - http://restingstate.com/

Resting state refers to the brain activity when we are not task driven. The major brain network when the brain is at rest is the Default Mode Network (DMN). This is associated with introspection. When this network is not functioning the way it is supposed to be – the result is a psychiatric pathology. http://en.wikipedia.org/wiki/Default_mode_network

Besides the DMN, there are the Dorsal Attention Network, the Visual Attention Network, the Salience Network. There is not strong agreement on what the other networks are. A signal processing technique independent component analysis (ICA) allows you to specify how many different networks you want to create and gives a quality number.

The Human Connectome Project (HCP) just released 500 individuals worth of high res images that have been heavily processed. - http://www.humanconnectome.com/
- Brain Genomics Superstruct Project - 1500 datasets - https://thedata.harvard.edu/dvn/dv/GSP
- indi 1000 connectomes http://fcon_1000.projects.nitrc.org/

Surface based “parcellation”, segmentation of the cortex was the only way that was presented. The neocortex is a 2D parametric surface folded up into a 3D space. FSL (freesurfer) and been incorporated in to the HCP toolset. http://freesurfer.net/

 
Imaging – I’m not really sure which of these are new, so, I am just listing them here – in order of how often they were mentions.
=======
7T MRI - 7 Tesla MRI machines from Seimens and GE.
Multislice MRI
QBall - 134 scan directions to be able to characterize whit matter tracts that cross each other. Better than DTI
fcMRI => functional connectivity
rs-fMRI => resting state functional MRI
MREG - MR Encephlography, 10 scans per second. a bit lower resolution
ASL -  arterial spin labelling (ASL)
FLAIR - Fluid-attenuated inversion recovery (FLAIR)
CSD tractography vs. ellipsoid. Better than DTI
 
Potential Clinical uses of brain imaging.
========================================
- early detection of Alzheimer’s
- early detection of schizophrenia
- predicting brain trauma / stroke outcomes and therapies
- characterize depression to advice treatment options
- neurofeedback
- autism, bipolar, anxiety disorders, ADHD, PTSD
** some are closer, more realistic than others.
fMRI and DTI have advised surgical planning for over a decade.

I was able to catch most of what was being said because of the fMRI & DTI Visualization work I did many years ago and because I have read at least a dozen books on consciousness, including 3 this year. image RestingStateNetworks450.jpg

Summary

The term Resting State was coined in 1995* that showed that many regions are active when the brain is resting, as opposed to having an active task. This led to the observation that there were many different networks of regions that were active at different times. The exact networks involved are not fixed, but the Default Mode Network (DMN), the Dorsal Attention Network, the Salience Network seem to be consistent, but, there seem to be about 10 +/- 3.

To the left are 13 networks determined by Independent Component Analysis (ICA), which takes reams of connectivity data and magically delineates. Paper and full image

The DMN is active when we let our minds wander introspectively. A strong DMN predicts better abilities at many other tasks. A disturbed DMN is associated with many mental illnesses. The DMN can be strengthened by mindfulness training. So, if the DMN is active when we are introspective, then the Oracle of Delphi's advice to "Know Thyself" is now backed up by the latest neuroscience.

Imaging

MREG, Based on highly undersampled data acquisition with inverse image reconstruction MR-Encephalography (MREG) allows to acquire whole-brain 3D-fMRI with data matrix of 64x64x64 at frame rates of 10 FPS. Kiviniemi show an amazing video showing a "neuronal avalanche".

http://www.humanbrainmapping.org/files/2012MeetingFiles/Sessions/Ultra-high%20speed%20fMRI%20Methods,%20Sensitivity%20Increases%20and%20Applications.pdf

grayordinates

Visualization

http://www.frontiersin.org/files/Articles/13956/fnana-05-00065-HTML/image_m/fnana-05-00065-g001.jpg

Random Notes

ex vivo - super high res slice by slice.

7T MRI gets down to 0.6mm isotropic voxels. - 170 slices per second.
- some skepticism about whether 7T will be useful in clinil setting since you need to be extremely still and just blood flow and breathing need movement correction. But, the tools seem to be getting better.

Pattern of convolution, tension based barin folding. The back of the brain is almost always the same between patients, but, the frontal cortex is much more individualized. That is why a surface segmentation is needed.

Lots of datasets are now available:
- The Human Connectome Project (HCP) has 500 high res images that have been heavily processed. - http://www.humanconnectome.com/
+ http://humanconnectome.org/documentation/S500/HCP500_GroupICA+NodeTS+Netmats_Summary_28aug2014.pdf
- Brain Genomics Superstruct Project - 1500 datasets - https://thedata.harvard.edu/dvn/dv/GSP
- indi 1000 connectomes http://fcon_1000.projects.nitrc.org/

HCP releasing software and processing pipeline and new data type "CIFTI" with grayordinates - http://www.nitrc.org/projects/cifti

FACT - Functional, Architecture, Connectivity, Topography - different to parcellate brain regions. How many parcels are there? Mouse has 40 areas.
130-140, 150-200, 55 areas???
http://en.wikipedia.org/wiki/Brodmann_area

MGH/Seimens working together on multislice MRI

QBall - instead of DTI - show fibers crossing. mulitshell QBall - which shells to choose. 60 directions, 134 directions, DQI?

ICA - In signal processing, independent component analysis (ICA) is a computational method for separating a multivariate signal into additive subcomponents.
- http://en.wikipedia.org/wiki/Independent_component_analysis
- basically, you throw all of the connectivity data you have into a box and tell it how many components you want and it makes them.
- this explains why there is a lot of variation in the number of different networks.

Cerebellum has 2 and probably 3 homotopic representations of the body used in processing sensory information.

Structural, Functional and effective connectivity are different.

Friday.
=======


Rich Club nuerons - hubs within brain regions - that take most of the long distance axons - layer 3 pyramid neurons
Flipping V1 - flips to V2 - reflections out from motor area - reflected connectivity
Connexels - 5D space - XYZ of edn points + connection graph
braingl - https://code.google.com/p/braingl/
progression of functional connectivity - 6 step progression - reflected connectivity
Connection Graphs galore.
Cortical expansion - geodetic folding - folds in occipital lobe are standard, less so in temporal, non-standard in frontal lobe. Cortex grows out from
hippocampus and orbital region and converges in area 46.
Modulation of connectivity - two regions talk to each other but a third region can inhibit/suppress it. P300 wave.
grayordinates, 178 parcels, gradients of connectivity?
spinal cord gray matter connectivity, inside the long haul white matter. strong local connectivity within slice and to nearest slices, some
connectivity to second closest slice, and then none.
fMRI on white matter - it is possible. CSD tractography vs. ellipsoid.
10 ms point ot point axon timing.

Vesa Kiviniemi
MREG - 10 scans per second. a bit lower resolution, different protocol.
- can see pulse and breathing.
- can see "neuronal avalanche" spread from sensory to full cortex.
Opening Blood Brain Barrier to let biologics thru to fight lymphoma


dFNC - dynamic functional connectivity - how the brain transitions from state to state.
meta states - sometimes multiple networks are active at the same time.
behavioral phenotype - genetics - brain imaging.
Glutamate - too much == manic, too little == depressed
imaging glutamate, imaging GABA and the connectivity between regions. correlate to which drugs are most useful.


stroke -> neglect - white matter
MY PEOPLE: everyone mentions voxels at least once.

James Hyde - "Giant Ugly 1mm Voxels"
One blood vessel per column -
200 micon cube - 0.2mm x 0.2mm x 0.2mm = 0.008 mm^3 cube - http://en.wikipedia.org/wiki/Vibratome
axon thickness 1-15 microns, aka, micrometers


the higher the connectivity between the central cortical nucleus of the amygdala and the dmPFC and dlPCF, the less anxiety,
dorsolateral prefrontal cortex (Brodmann area 46)

Last talk on Friday - everything you know is WRONG. Van Weedan
- white matter is in sheets as 2D mesh or grids that intersect and form a grid.
- white matter is an active participate in processing - noisy amplifier

Saturday
========
VIM nucleus - http://www.ncbi.nlm.nih.gov/pubmed/14557560 Deep Brain Stimulation
MR-guided Focused Ultrasound Thalamotomy for Essential Tremor - thermal ablation.

Formation of the DMN - prenatal early opposition of the PCC and other brain areas + PCC and subcortial
Neurocognitive Network
language tens to be left lateral, with hemispherectomy - taking out 1/2 of brain to alleviate seizures, plasticity language can migrate
Ideal age for brain function recovery seems to be age 9.

Resting State imaging can predict intelligence!
Sleep deprivation and brain networks - Ventral Attention, Dorsal Attention

Psychoradiology
PTSD - reduction in brain size
cortical thickness in medial cortex increases in schizophrenia
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3997602/

fMRI for neurofeedback therapy

ADHD reaction time variation - ventral striation and surface area.

Open Science. so many datasets. "Connectome in a Box is like XMAS in August", "This is a watershed moment"

Thomas Insel - NIMH crosslink genes, molecules, neural circuits, physiology, behavior and self informing.
- morbidity of coronary, cancer have dropped by a lot. Suicide morbidity is on the rise. 90% of suicide is mental illness
- psychiatric illness is the most costly to the economy of all illnesses.
+ http://www.nimh.nih.gov/about/director/2011/the-global-cost-of-mental-illness.shtml
Who will pay for imaging?
Brain 2025


Mental Illness is aa arythmia, not an infarction. "connectopathies"
Tyrany of symptoms, DSM4
RDoC - Research Domain Criteria project (RDoC) - http://www.nimh.nih.gov/research-priorities/rdoc/index.shtml
Be able to identify mental illness by biomarkers, the way there are 3 identifiable subtypes of breast cancer.

synaptic failure, neurodegeneration, cortical thickness
Deep Brain Stimulation (DBS) - 10Hz excitatory, 1Hz inhibitory
less than 40% of depression responds to a single drug. trial and error.
Drugs, DBS, CBT - Cognitive Behavioral Therapy - together get to 72% of patients.

Introspective Awareness - Mindfulness Training strengthens DMN

--
http://www.nih.gov/about/director/06052014_statement_brain.htm - Statement on the Long-Term Scientific Vision for the BRAIN Initiative
http://www.nih.gov/science/brain/2025/ - Brain 2025 Vision Stement

unrelated -
http://www.neurobureau.org/galleries/brain-art-competition-2011/
http://www.neurobureau.org/galleries/brain-art-competition-2012/
http://www.neurobureau.org/galleries/brain-art-competition-2013/
http://www.neurobureau.org/galleries/brain-art-competition-2014/
http://www.neurobureau.org/wp-content/uploads/2013/03/41.png
http://fieldtrip.fcdonders.nl/faq/how_are_the_different_head_and_mri_coordinate_systems_defined
https://github.com/the-virtual-brain/tvb-library/wiki/A-Description-of-a-Complete-Dataset
http://cercor.oxfordjournals.org/content/early/2013/01/01/cercor.bhs397.short
http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/Atlases
http://www.bic.mni.mcgill.ca/ServicesAtlases/HomePage


Visualization:
Connectivity Matrix: http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0016322
DTI - structural connectivity
Functional Brain Network - http://www.nature.com/nrn/journal/v10/n3/box/nrn2575_BX1.html
circular display of connectivity - http://www.humanconnectomeproject.org/informatics/relationship-viewer/
[ TO DO !!! ]

http://www.intechopen.com/source/html/41878/media/image1.png - Diffusion Tensor Tractography (left), graph of connectivity (center), functional connectivity (right).

Potential Cinical uses of brain imaging. ========================================
- early detection of Alzheimers
- early detection of schizophrenia
- predicting brain trauma / stroke outcomes and therapies
- characterize depression to advice treatment options
- neurofeedback
- autism, bipolar, anxiety disorders, ADHD, PTSD
** some are closer, more realistic than others.
fMRI and DTI have advised surgical planning for over a decade.

DMN
DMN & psych pathologies - http://neurowiki2012.wikispaces.com/Default+Mode+Network http://en.wikipedia.org/wiki/Default_mode_network

As Silver/Gold sponsors Seimens & GE got to make 20 minute presentations right before lunch on Thursday and Friday. Seimens especially is investing heavily in fMRI, it seems. both companies have introduced 7T scanners.

---------------------------------- ASL arterial spin labeling (ASL), which allows the weighting of the MRI signal by cerebral blood flow.

----------------------------------

Posters
Rui Yuan - Cortical Thalamic network
Melanie Pincus - Sacred values and decision making
Drew Lindsey - Scene Complexity and task based connectivity
Anna Custo - microstates
Idan Blank - language congnitive control
J. Athilingam - TPJ - N250 - social stimuli

I think TC let me go to this in part because I spent so, so, so long debugging the VP2000 in R720 - the most boring task in my professional career. And because I have not been out of the office in years and years.

I was able to catch most of what was being said because of the fMRI & DTI Visualzation work I did many years ago: http://jch.com/volumes/dtifmri.htm And because I have read at least a dozen books on consciousness, including 3 this year. - http://www.jch.com/jch/notes/Consciousness.html

*1995 Paper: Functional connectivity in the motor cortex of resting human brain using echo-planar mri Bharat Biswal, F. Zerrin Yetkin, Victor M. Haughton and James S. Hyde Ph.D.*

image badge.jpg
Cap: badge.jpg
image raw0.jpg
Cap: raw0.jpg
image raw1.jpg
Cap: raw1.jpg
image raw2.jpg
Cap: raw2.jpg
image raw3.jpg
Cap: raw3.jpg
image raw4.jpg
Cap: raw4.jpg
2014.09.13 jch