Simon Eickhoff

@INM7_ISN

News & Views on Brain mapping | Machine-Learning | Translational Neuroscience. Director INM-7 ; Professor for Systems Neuroscience .

Joined December 2017

Tweets

You blocked @INM7_ISN

Are you sure you want to view these Tweets? Viewing Tweets won't unblock @INM7_ISN

  1. Pinned Tweet
    2 May 2019

    Finally, our a new website highlighting research & people Infos on the groups and their PIs (-> Research) Links to publicly shared tools / data (-> Resources) Media information and archive (-> Publications & Media)

    , , and 7 others
    Undo
  2. Retweeted
    17 hours ago

    Informative and entertaining read on how counter-intuitive high dimensional space is, on the improbable effectiveness of and the concept of The easy answer to the unreasonable effectiveness, though? overfitting

    Undo
  3. 17 hours ago

    Informative and entertaining read on how counter-intuitive high dimensional space is, on the improbable effectiveness of and the concept of The easy answer to the unreasonable effectiveness, though? overfitting

    Undo
  4. Retweeted
    Jan 17

    The "pain-matrix" in the human brain revealed by our new meta-analysis: Preregistered, ~5000 screened papers, >200 included experiments Core regions consistently recruited regardless of stimulation type, pain location & participant sex

    Undo
  5. Retweeted
    Jan 31

    Happy to advertise a 3yr EU-funded PhD position in Lille (co-supervised w ) on based of More on application very soon

    Undo
  6. Retweeted
    Jan 29

    Heute hat Edna Cieslik-Köchling den Schülern des Gymnasiums beim Hirnforschungstag im einen Einblick in die Hirnforschung und unseren Arbeitsbereich der interindividuellen Variabilität des Gehirns gegeben. Eine schöne Möglichkeit, um von Experten zu lernen!

    Undo
  7. Retweeted
    Jan 24
    Undo
  8. Retweeted

    Delighted to have Prof. Simon Eickhoff (Institute of Systems Neuroscience, U. of Dusseldorf) visit the , & give a talk at today’s on: “Combining brain mapping with machine-learning on individual differences”

    Undo
  9. Retweeted
    Jan 23

    Sumo wrestlers and basketball players ⛹️‍♀️ may be statistically indistinguishable (p=0.1) if you compare their weights or height, but multivariable approach (height&weight) has >90% classification accuracy. This is an essential lesson for psychiatric neuroimaging!

    Undo
  10. Jan 22

    “Using a term like nonlinear function is like referring to the bulk of zoology as the study of non-elephant animals.” – Stanislaw Ulam

    Undo
  11. Retweeted

    The finding of common neurocircuit substrates for the cognitive and emotional impairment prevalent across disorders could advance multimodal therapeutics and reconceptualize mental illness.

    Brain
    Undo
  12. Jan 22

    Given publication delays, articles lag behind the actual work by month if not years. So when evaluating applications for early / mid-career researchers, which would you consider in your assessment?

    Undo
  13. Jan 20

    Me, trying to explain heritable traits to my kids: me: "what's your hair color?" K: "brown" me: "and what's my hair color?" K: "grey!"

    Undo
  14. Jan 20

    Dimensions of schizophrenia psychopathology by 4 factor provide the most stable & generalizable representation based on ~2000 patients from 10+ sites: Here is our new webtool to map data onto these dimensions !!

    Undo
  15. Jan 19

    To be useful, on neuroimaging data needs to generalize across centers, but site specific effects are massive and a major challenge Excited to see this work by & team using ComBat for harmonization across datasets 6 lifespan

    Undo
  16. Jan 19

    Is anybody aware of a (single) pre-registered neuroimaging machine-learning study? That is, classifier optimized before code is independently deposited and study registered. New data acquired and algorithm run once. Does any such paper exist?

    Undo
  17. Jan 18

    Patients with show reproducible patterns of functional connectivity: by Or not? : stabilities across pipelines but not sites Problem: The samples in the second negative study are an order of magnitude smaller

    Undo
  18. Jan 18

    Can you change your personality? Apparently: The idea of stable personality always struck me odd as plasticity is probably *the* defining feature of the brain. Why should (only) personality be unaffected by experience or willful effort?

    Undo
  19. Retweeted
    Jan 17
    Undo
  20. Jan 17

    Double Dipping in yields 80% accuracy on random data but is easily prevented by clean crossvalidation Bigger issue: overfitting by re-re-re-analysing datasets. Data-sharing will help, but ultimately it's prospective, prepregisterd tests

    Undo
  21. Jan 17

    The "pain-matrix" in the human brain revealed by our new meta-analysis: Preregistered, ~5000 screened papers, >200 included experiments Core regions consistently recruited regardless of stimulation type, pain location & participant sex

    Undo

Loading seems to be taking a while.

Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.

    You may also like

    ·