Thomas Yeo @bttyeo Twitter profile | Pikagi

Pikagi

Thomas Yeo

@bttyeo

11,744

Followers

1,380

Following

164

Media

12,125

Statuses

brain imaging, machine learning, neuroscience, mental disorders

https://t.co/CFZhvFIuCt

Joined July 2016

Don't wanna be here? Send us removal request.

Pinned Tweet

@bttyeo

Thomas Yeo

3 months

1/11 The poll has ended: 53% (scan time) vs 47% (sample size)! Here's our in-depth take on brain-wide association studies (with a 134-page supplement!): Led by @Leon_Oo1 @csabaorban , there are a few twists, so do read till the end! Our previous study ...

Tweet media one

@bttyeo

Thomas Yeo

3 months

Given a fixed budget, should we prioritize sample size or scan time of each participant for resting-state fMRI?

19

16

32

5

110

252

Last Seen Profiles

@Tsm_Ryu

@DanielMorgen87

@ohahpotato

@pico_staff_H

@AnnFink55

@stw_pdg

@Marcioshalon777

@Hania_KHan_

@Lilahvtube10342

@AndaLookkaew_CN

@sergioh17699374

@alphat3hellborn

@not0tomorrow

@NazFX_Studios

@aylulmadrid

@alexandraocamp1

@GWimhurst

@cineanimeytv

@Spartn_

@yargi_evrenii

@0x_Azuki

@LynnGilmartin

@Aether23148

@cybuniverse

@FrankFigueired0

@MugiwaraArt

@_MorrTex

@artonavax

@coachcollins92

@jefftheironicda

@LibertyMSoccer

@KnightsEditing

@mstrubell_EN

@BSCS_Global

@LFSMx

@CapitalMX_

@bttyeo

Thomas Yeo

4 years

It’s official! I got tenure! Time to air my grievances! 1/52

155

47

3K

@bttyeo

Thomas Yeo

1 year

Unpopular opinion: it should be ok to self plagiarize the data description and preprocessing portions of papers.

47

31

931

@bttyeo

Thomas Yeo

1 year

I have resigned as senior editor of NeuroImage. The publication fee is unjustifiably high and I can no longer support the journal in good conscience. We hope everyone will support the birth of our new journal Imaging Neuroscience!

@ImagingNeurosci

Imaging Neuroscience EiC

@ImagingNeurosci

1 year

All NeuroImage and NeuroImage:Reports editors have resigned over the high publication fee, and are starting a new non-profit journal This comes with great regret, and a huge amount of thought and discussion- please read announcement to get more details.

Tweet media one

Tweet media two

170

2K

5K

8

67

649

@bttyeo

Thomas Yeo

3 years

Come on google scholar! I just want to find 11780 citations!

11

29

622

@bttyeo

Thomas Yeo

4 years

I am looking for a postdoc with PhD training in neuroscience, psychology or related field. Topics are quite flexible () and up for discussion. Appreciate RT.

6

305

300

@bttyeo

Thomas Yeo

3 years

I am looking for a postdoc with PhD training in neuroscience, machine learning or related field. Topics are quite flexible () and up for discussion. Appreciate RT.

5

227

288

@bttyeo

Thomas Yeo

3 years

1/5 Paper is out: We found multiple complete "representations" of the cerebral cortex within individuals' cerebellum, including two distinct default networks. @xueaihuiping @rubykong92 @LaurenDiNicola @RodBraga @pangeli95

Tweet card media

The detailed organization of the human cerebellum estimated by intrinsic functional connectivity...

Distinct regions of the cerebellum connect to separate regions of the cerebral cortex forming a complex topography. Although cerebellar organization has been examined in group-averaged data, study of...

journals.physiology.org

@WiringTheBrain

Kevin Mitchell

@WiringTheBrain

4 years

The Detailed Organization of the Human Cerebellum Estimated by Intrinsic Functional Connectivity Within the Individual - beautiful work from @bttyeo + colleagues detailing connectivity b/w cerebellum + cortex

Tweet media one

Tweet media two

2

58

144

7

92

256

@bttyeo

Thomas Yeo

4 years

What happens to the brain's functional network organization if we remove one cortical hemisphere? Excited to be involved in this study spearheaded by Dorit Kliemann: With @neurexplorer

Tweet media one

5

87

233

@bttyeo

Thomas Yeo

3 years

1/3 Our study on individual-specific areal-level parcellations is out Multi-resolution HCP parcellations & FC matrices found here Spearheaded by @rubykong92 @gordonneuro @zuoxinian @Nathan_Spreng @getian107 @AvramHolmes @INM7_ISN

Tweet media one

3

89

233

@bttyeo

Thomas Yeo

2 years

I am looking for a postdoc to start this year or 2023. Candidate should have a PhD in neuroscience, machine learning or related fields. Topics are quite flexible () and up for discussion. Appreciate RT.

Computational Brain Imaging Group - Career

We are looking for a research assistant/engineer with a bachelor's or master's degree in engineering, computing, physics, mathematics or related subject. Interested candidates can email their CV to...

sites.google.com

2

191

227

@bttyeo

Thomas Yeo

5 years

Honored to receive the award. I hope to live up to it.

@INM7_ISN

Simon Eickhoff

5 years

Extremely well deserved. Couldn't think of a more deserving @ohbm #OHBM2019 early career award winner. Congratulations @bttyeo

6

5

109

21

2

225

@bttyeo

Thomas Yeo

3 years

I am an editor of a journal which has desk rejected all of my papers (so far).

@SarahLWhittle

Sarah Whittle

3 years

Anyone else constantly asked to review papers at journals that have never published your work (i.e., have rejected every paper you've ever submitted to them)?

17

3

149

7

6

202

@bttyeo

Thomas Yeo

5 years

For those who are interested, here's the slides from my talk on Generative Modeling of Resting-State Parcellations & Dynamics at #OHBM2019

1

64

199

@bttyeo

Thomas Yeo

1 year

TFW my resignation gets more attention than my papers on Twitter.

6

1

195

@bttyeo

Thomas Yeo

3 years

1/12 Our new preprint develops a spatially heterogeneous biophysical model to explain functional connectivity dynamics. With @DoctorKong @csabaorban @rubykong92 @AvramHolmes @KevinMAnder @martijnheuv @ZShaoshi @johndmurray @pengwanghome & Gustavo Deco

Tweet media one

2

64

185

@bttyeo

Thomas Yeo

6 months

The publication fees for HBM have dropped from $3850 to $3200. I might be presumptuous here but I believe this is the success of our collective resignation from the two NeuroImage journals and the establishment of Imaging Neuroscience as a high quality lower cost alternative.

Tweet media one

5

16

175

@bttyeo

Thomas Yeo

2 years

Resting-state fMRI has finally shown clinical utility!

@Galen_CL_Hung

Galen Hung

2 years

SAINT from Stanford has just received FDA approval as an accelerated and individualized TMS for depression!

0

12

54

4

23

170

@bttyeo

Thomas Yeo

4 years

1/5 Our new preprint uses meta-learning to boost functional connectivity based prediction of *NEW* phenotypes in small samples. We achieve 100%-400% improvements in many scenarios! @hetong93 @AvramHolmes @anlijuncn @danilobzdok @AvramHolmes @INM7_ISN

7

65

162

@bttyeo

Thomas Yeo

3 years

Code available upon request

@omarsar0

elvis

3 years

machine learning researcher's codebase

28

192

925

1

26

164

@bttyeo

Thomas Yeo

5 years

“This suggests that DL - in fact - does not improve prediction compared to simpler baseline models.“ - controversy about deep learning paper predicting earthquake published in Nature.

Tweet card media

One neuron is more informative than a deep neural network for...

29 August 2018: "Artificial intelligence nails predictions of earthquake aftershocks". This Nature News headline is based on the results of DeVries et al. (2018) who forecasted the spatial...

3

39

156

@bttyeo

Thomas Yeo

1 year

I have RESIGNED from the Schaefer parcellation and started a new homotopic parcellation!

@mathilde0427

Xiaoxuan (Matilda) Yan

1 year

Paper officially out! Check it out here: Check out our GitHub repo for our parcellations and more:

Tweet media one

2

34

123

5

16

154

@bttyeo

Thomas Yeo

5 years

@70Hertz LOL. So by their standards, being the 25th author on a 50-author paper in Nature Comms is somehow better than being a first author in a paper published in a “lower-profile” journal.

2

3

151

@bttyeo

Thomas Yeo

5 years

1/6 Preprint led by @pretty_Jingwei @rubykong92 with @AvramHolmes @getian107 @mertrory @R_Liegeois @csabaorban Global signal regression (GSR) is highly controversial: it's great at removing global artifacts, but introduces negative correlations, etc

Tweet media one

Tweet media two

3

81

149

@bttyeo

Thomas Yeo

2 months

Excited our first patient with treatment resistant depression has enjoyed remission after 1 week of intensive TMS with personalized targeting from @rubykong92 algorithm! Congrats to clinical team led by Dr @Torphernchern and also @xueaihuiping Trevor Tan @Leon_Oo1

2

14

142

@bttyeo

Thomas Yeo

4 years

Interested in individual-specific cerebellar network organization? Check out @xueaihuiping poster #1148 at #OHBM2020 , where she studied 2 individuals with 31 sessions (62 runs) of data. #OHBM2020 @OHBM Joint work with @rubykong92 , Randy Buckner & others.

Tweet media one

3

34

139

@bttyeo

Thomas Yeo

3 years

Excited study is now published! Free access Code Led by @DoctorKong w/ @csabaorban @rubykong92 @AvramHolmes @KevinMAnder @martijnheuv @ZShaoshi @johndmurray @pengwanghome Gustavo Deco @NatureComms

Tweet card media

Sensory-motor cortices shape functional connectivity dynamics in the human brain

Nature Communications - Spontaneous fluctuations in brain activity exhibit complex spatiotemporal patterns across animal species. Here the authors show that sensory-motor regions and spatial...

@bttyeo

Thomas Yeo

3 years

1/12 Our new preprint develops a spatially heterogeneous biophysical model to explain functional connectivity dynamics. With @DoctorKong @csabaorban @rubykong92 @AvramHolmes @KevinMAnder @martijnheuv @ZShaoshi @johndmurray @pengwanghome & Gustavo Deco

Tweet media one

2

64

185

4

53

137

@bttyeo

Thomas Yeo

4 years

Will academic publishers do the same thing?

@Complex

Complex

4 years

PornHub has now made its premium services free worldwide, and will also donate to the #Coronavirus relief efforts:

Tweet media one

714

4K

13K

4

19

130

@bttyeo

Thomas Yeo

5 years

Here’s my lab’s github: Contains some python code but mostly matlab code. Enjoy!

Tweet card media

GitHub - ThomasYeoLab/CBIG

Contribute to ThomasYeoLab/CBIG development by creating an account on GitHub.

2

8

129

@bttyeo

Thomas Yeo

6 years

Trying to map between volumetric (MNI152) and surface (fsaverage) coordinate systems? See our new preprint Test drive the code here: with @Jadecci @INM7_ISN @rubykong92

Tweet media one

Tweet media two

Tweet media three

Tweet media four

2

63

127

@bttyeo

Thomas Yeo

4 years

Having trouble interpreting predictive models? Do predictive network features differ across behaviors and/or brain states? Check out @JianzhongChen9 & @angelatamtweets poster #0090 #OHBM2020 #OHBM2020Posters with @danilobzdok @INM7_ISN @AvramHolmes @ndosenbach @valeria_kebets

Tweet media one

3

41

127

@bttyeo

Thomas Yeo

2 years

1/5 Quite a number of posters from our group today! #OHBM2022 First, if you have been using the Schaefer parcellation, check out @mathilde0427 poster WTh618, where we develop a homotopic version of the Schaefer parcellation. This will be great for lateralization studies!

Tweet media one

4

20

125

@bttyeo

Thomas Yeo

5 years

New paper with Peng Wang, Gustavo Deco @martijnheuv , @rubykong92 @R_Liegeois @csabaorban @scienceadvances Most neural mass models assume each brain region has same micro-scale parameters. We relax this assumption & automatically estimate the parameters.

3

82

126

@bttyeo

Thomas Yeo

6 years

1/6 Thread on new work. Free access: (1) Our method greatly improves individuals' parcellations (2) Individual-specific network topography can predict behavior [cont…] @rubykong92 @AlexandSchaefer @zuoxinian @INM7_ISN @mertrory @AvramHolmes @csabaorban

Tweet media one

2

75

122

@bttyeo

Thomas Yeo

2 years

1/9 Paper is out Here's some takeaways including new tidbits (good & bad!) since the preprint & even after the paper was accepted. Large sample sizes are helpful for individual-level prediction. We have known that for quite some time, but the recent...

@bttyeo

Thomas Yeo

4 years

1/5 Our new preprint uses meta-learning to boost functional connectivity based prediction of *NEW* phenotypes in small samples. We achieve 100%-400% improvements in many scenarios! @hetong93 @AvramHolmes @anlijuncn @danilobzdok @AvramHolmes @INM7_ISN

7

65

162

2

44

120

@bttyeo

Thomas Yeo

5 years

1/4 New preprint () with @tangsy93 @SunNanbo @DoriFloris & Adriana Di Martino. Autism is well-known to be heterogeneous, but (categorical) subtypes miss out on continuous (dimensional) inter-individual variation. So we used a Bayesian model to ...

Tweet media one

4

60

120

@bttyeo

Thomas Yeo

1 year

I believe this new parcellation will be super useful not just for lateralization studies, but also brain imaging studies in general. We are already using it for neural mass modeling!

@mathilde0427

Xiaoxuan (Matilda) Yan

1 year

Paper officially out! Check it out here: Check out our GitHub repo for our parcellations and more:

Tweet media one

2

34

123

1

26

115

@bttyeo

Thomas Yeo

6 years

Want to map group activation or meta-analytic maps from MNI to fsaverage? Want to map fsaverage parcellations to MNI? Want to map Julich architectonic areas from Colin27 to fsaverage? Our paper is out Preprint Congrats to @Jadecci

Tweet media one

Tweet media two

Tweet media three

Tweet media four

1

56

115

@bttyeo

Thomas Yeo

4 years

I am not a fan of this new trend of posters containing many pretty pictures but minimal methodological details. A poster shouldn’t have too many words but there should be enough words so audience can follow roughly what was done. #OHBM2020 #OHBM2020Posters

5

3

112

@bttyeo

Thomas Yeo

4 years

1/x For those interested, here's the dropbox link to the parcellation: A tweet thread counts as a publication nowadays right? So let me highlight some cool insights we found.

Tweet card media

ThalamusParcellation.zip

Shared with Dropbox

www.dropbox.com

@bttyeo

Thomas Yeo

4 years

I have a thalamic parcellation from 2010 that is still unpublished. I have been sharing it with everyone who asks for it, so at least it’s not wasted.

9

5

63

6

38

111

@bttyeo

Thomas Yeo

5 years

Thanks to feedback from 8 different reviewers (!), our paper on latent factors underlying atrophy, cognitive and tau heterogeneity in AD dementia is finally out in @NeuroImage_EiC Free access (for 50 days): @SunNanbo @mertrory Beth Mormino, Jianzhong Chen

@bttyeo

Thomas Yeo

6 years

We found 3 atrophy patterns that predict memory, language & executive function in late-onset AD. Results replicate in 2nd dataset. Atrophy-cognitive patterns were correlated with PET Tau spatial patterns in MCI participants @SunNanbo Beth Mormino @mertrory

Tweet media one

Tweet media two

Tweet media three

Tweet media four

0

39

90

1

28

112

@bttyeo

Thomas Yeo

4 years

As 2010s come to an end, I am grateful for a happy family & getting paid to do what I love. But I hope humanity will finally take urgent climate action next decade. There is little time left. We might even be past the tipping point. But we cannot and must not give up.

0

7

110

@bttyeo

Thomas Yeo

4 years

Isn’t it brave of Boris to volunteer for his herd immunity program?

3

5

108

@bttyeo

Thomas Yeo

2 years

Happy lunar new year everyone!

3

4

101

@bttyeo

Thomas Yeo

4 months

Grant writing is going really well!

Tweet media one

7

2

101

@bttyeo

Thomas Yeo

5 years

1/3 New preprint () showing that Somatosensory-Motor Functional Dysconnectivity Spans Multiple Transdiagnostic Dimensions of Psychopathology. Led by @valeria_kebets with @AvramHolmes @russpoldrack @pretty_Jingwei @rubykong92 @csabaorban @tangsy93

Tweet media one

5

57

99

@bttyeo

Thomas Yeo

6 years

Revised paper (1) Individual-specific network location/topography predicts 58 cognitive, emotion & personality measures. (2) Network topography better than network size for prediction. (3) Our method comparable to two popular methods using 5X the data.

Tweet media one

Tweet media two

Tweet media three

2

68

97

@bttyeo

Thomas Yeo

4 years

COVID-19 is a like an easier trial run of the climate crisis. Yet many countries have failed the test spectacularly.

1

16

98

@bttyeo

Thomas Yeo

5 years

Preprint: We compare kernel regression with deep neural networks (DNNs) for functional connectivity behavioral prediction using ~10,000 participants. Hint: DNNs do not perform better @hetong93 @rubykong92 @danilobzdok @INM7_ISN @mertrory @AvramHolmes

Tweet media one

Tweet media two

Tweet media three

Tweet media four

5

46

95

@bttyeo

Thomas Yeo

6 years

Dear Journals, please do not force reviewers to log in when inviting them to respond to review requests. Those requests will simply not be answered.

3

21

95

@bttyeo

Thomas Yeo

1 year

Super excited to handle my first paper for Imaging Neuroscience!!

@ImagingNeurosci

Imaging Neuroscience EiC

@ImagingNeurosci

1 year

We are now open for consideration of new papers at Imaging Neuroscience: This allows for the review process to proceed without delay. The production & publication process for accepted papers will then wait until full journal opening (aiming for mid-July).

6

225

479

0

3

93

@bttyeo

Thomas Yeo

3 years

I would not be here without my PhD supervisor Professor Polina Golland. I would do my PhD again with her in a heartbeat.

1

5

92

@bttyeo

Thomas Yeo

5 years

New preprint with Minh Nguyen, @hetong93 & others: We propose a RNN to predict Alzheimer's disease progression up to 6 years into the future. Our model is currently 2nd on the TADPOLE leaderboard:

Tweet media one

3

26

92

@bttyeo

Thomas Yeo

4 years

Machine learning requires big data. In neuroimaging, there are big datasets, e.g. UK Biobank. Can big data help prediction in smaller data with *different* behaviors? Check out simple method by @hetong93 at poster #1095 #OHBM #OHBM2020 with @danilobzdok @INM7_ISN @anlijuncn

Tweet media one

2

35

90

@bttyeo

Thomas Yeo

6 years

We found 3 atrophy patterns that predict memory, language & executive function in late-onset AD. Results replicate in 2nd dataset. Atrophy-cognitive patterns were correlated with PET Tau spatial patterns in MCI participants @SunNanbo Beth Mormino @mertrory

Tweet media one

Tweet media two

Tweet media three

Tweet media four

0

39

90

@bttyeo

Thomas Yeo

2 months

Beautiful study by @jingnandu049 @LaurenDiNicola Peter Angeli and Randy Buckner on individualized network topography with minor involvement by me. This study uses the multi-session hierarchical Bayesian model developed by @rubykong92

Tweet card media

Organization of the Human Cerebral Cortex Estimated Within Individuals: Networks, Global Topograp...

The cerebral cortex is populated by specialized regions that are organized into networks. Here we estimated networks from functional MRI (fMRI) data in intensively sampled participants. The procedure...

journals.physiology.org

1

22

90

@bttyeo

Thomas Yeo

4 years

Gradients🌈, soft parcellations & hard parcellations can be used to predict behavior. How well do they do? Check out @rubykong92 poster #1075 at #OHBM2020 #OHBM2020Posters with @BorisBernhardt , @INM7_ISN & others not on twitter.

Tweet media one

0

26

88

@bttyeo

Thomas Yeo

4 years

Yet another giant self-inflicted damage to the U.S. I was on a J1 visa when I was doing my PhD and postdoc in the U.S.

@NPR

NPR

4 years

NEW: President Trump is expected to sign an order as soon as Saturday to suspend H-1B, L-1 and other temporary work visas through the end of the year, according to the multiple sources familiar with the plan.

295

777

1K

5

15

88

@bttyeo

Thomas Yeo

3 years

This position has been filled. BUT I have opening for a second position. Topics are quite flexible () and up for discussion. Appreciate RT.

@bttyeo

Thomas Yeo

4 years

I am looking for a postdoc with PhD training in neuroscience, psychology or related field. Topics are quite flexible () and up for discussion. Appreciate RT.

6

305

300

1

101

87

@bttyeo

Thomas Yeo

4 years

Very impressed by the #OHBM2020 Open Science Room. Maybe @OHBM can pay the volunteers at @OhbmOpen to organize the next virtual OHBM for #OHBM2021 (Yes, I am pessimistic about the possibility of a physical OHBM next year.)

3

10

84

@bttyeo

Thomas Yeo

2 years

1/ Others have already mentioned that non-BWAS / non-human studies might not have similar issues, so I won't comment on that except to highlight @ten_photos point that this issue is not restricted to BWAS or human or fMRI or even brain studies:

@smarek0502

Scott Marek

2 years

A dream come true, now out @Nature w @tervoclemmensb I started this project before knowing my wife and kid. Science can be a slog, this one was worth it. Pointing towards an exciting future for brain-behavior research. A🧵👇

Tweet media one

Tweet media two

Tweet media three

8

134

593

1

11

82

@bttyeo

Thomas Yeo

5 years

Come see our lab’s #OHBM2019 posters! M068: @valeria_kebets shows her work on transdiagnostic hubs of Psychopathology from this preprint: M585: @rubykong92 shows her work on individual-specific areal-level parcellations 1/3

Tweet media one

Tweet media two

2

16

80

@bttyeo

Thomas Yeo

5 years

Dynamic mode decomposition of resting-state and task fMRI. Paper is out @NeuroImage_EiC with @R_Liegeois & @dvdevill Free version:

0

36

81

@bttyeo

Thomas Yeo

1 year

As authors/reviewers/editors, we *are* the market :) In our discussion with Elsevier, they claim that the current $3450 is justified by market forces. The market has spoken!

@Christakou

Anastasia Christakou 💙

1 year

Turns out market forces are not the only forces at play. Some things are priceless. Kudos to the NeuroImage+ editors for leading by example 👊

1

2

40

2

7

80

@bttyeo

Thomas Yeo

1 year

A loss to our @OHBM community, but a great gain for the ESA! Big congratulations to Raphaël @Raph_Astro @vdcalhoun , so is the time-varying connectivity educational course still going to happen? Is Raphaël going to do the educational course from space??

@esa

European Space Agency

1 year

Meet ESA's class of 2022 astronauts 👇 #ESAastro2022

Tweet media one

148

1K

5K

9

5

77

@bttyeo

Thomas Yeo

2 years

Does that mean I am no longer an early career researcher? 🤔

@DrBreaky

Michael Breakspear

2 years

My term as Editor-in-Chief of NeuroImage @NeuroImage_EiC concludes at the end of this month Delighted to announce that @fmrib_steve is the incoming Editor-in-Chief @bttyeo is stepping up to a Senior Editor post @emrobSci and @cately00 join NeuroImage as new Handling Editors

Tweet media one

23

15

441

3

2

77

@bttyeo

Thomas Yeo

5 years

@HirstRj We use the Schaefer atlas from our own group: It is multi-scale: 100 to 1000 parcels. Available in MNI & surface space. We have shown that it's great for fMRI data @INM7_ISN has shown that it also works well for VBM. Hope to release it in FreeSurfer soon.

Tweet media one

Tweet media two

7

15

76

@bttyeo

Thomas Yeo

4 years

Foolish of me to think I can focus on work today.

6

2

76

@bttyeo

Thomas Yeo

3 years

Exciting lineup for the Centre for Translational MR Research Symposium on 10 Nov 2021! First session is from 9am to 11am (Singapore time), which is evening for USA. Second session is from 3pm to 5pm (SGT), which is morning for Europe. You can sign up here

Tweet media one

Tweet media two

2

30

73

@bttyeo

Thomas Yeo

4 years

I love coding. Way more fun than administrative work.

@bayes_baes

bayes baes

4 years

how do people code for like 30+ yrs and claim to still love it this shit sucks stop lying

16

3

144

6

0

73

@bttyeo

Thomas Yeo

5 years

One problem with male toilets not having urinals is that I have to keep double checking whether I am in the wrong restrooms. #FirstWorldProblems #OHBM2019

5

1

73

@bttyeo

Thomas Yeo

4 years

Paper is out in @NeuroImage_EiC :

@bttyeo

Thomas Yeo

5 years

New preprint with Minh Nguyen, @hetong93 & others: We propose a RNN to predict Alzheimer's disease progression up to 6 years into the future. Our model is currently 2nd on the TADPOLE leaderboard:

Tweet media one

3

26

92

2

19

73

@bttyeo

Thomas Yeo

3 years

Finishing an epic revision with a 49-page single-spaced response letter. Congrats to @JianzhongChen9 and @angelatamtweets for breaking the lab's record.

2

0

71

@bttyeo

Thomas Yeo

2 years

Exciting keynote by @AndrewZalesky at #OHBM2022

Tweet media one

1

3

72

@bttyeo

Thomas Yeo

2 years

Smiley faces should be allowed in responses to reviewers.

4

0

70

@bttyeo

Thomas Yeo

8 months

Having great collaborators is not just the key to success but frankly one of the best features of academia!

@jayvanbavel

Jay Van Bavel, PhD

8 months

An underrated key to academic success: finding some badass collaborators who do great work, keep things fun, and can hit a deadline.

7

24

332

1

1

68

@bttyeo

Thomas Yeo

5 years

Paper is finally out in @NeuroImage_EiC : Code available here: @hetong93 @danilobzdok @INM7_ISN @mertrory @AvramHolmes @rubykong92

@bttyeo

Thomas Yeo

5 years

1/2 Updated preprint () showing that kernel regression and DNNs have equal performance for RSFC behavioral prediction. In this version, we found that ... @hetong93 @rubykong92 @danilobzdok @INM7_ISN @mertrory @AvramHolmes

1

23

51

0

25

66

@bttyeo

Thomas Yeo

6 years

New work with Maxwell Bertolero, @DaniSBassett and Mark D'Esposito. Diverse hubs associated with greater modularity and better task performance

Tweet media one

Tweet media two

1

46

64

@bttyeo

Thomas Yeo

11 months

Very excited about new study from our group!

@ZShaoshi

Zhang Shaoshi

11 months

1/11 Super excited to share our latest preprint! 📢 We estimate a putative marker of whole-cortex excitation/inhibition (E/I) ratio that indexes cortical maturation in youth! Leveraging a large-scale biophysically plausible model, we can estimate E/I ratio in-vivo! 🧵

Tweet media one

5

52

213

3

11

66

@bttyeo

Thomas Yeo

3 months

Super excited about this new preprint. Will do a thread on it later today.

@biorxiv_neursci

bioRxiv Neuroscience

@biorxiv_neursci

3 months

MRI economics: Balancing sample size and scan duration in brain wide association studies #biorxiv_neursci

0

21

55

2

18

65

@bttyeo

Thomas Yeo

1 year

Can GPT-4 submit a concur travel claim successfully with just one try?

4

4

65

@bttyeo

Thomas Yeo

5 years

Congrats to Dr. Ruby Kong @rubykong92 for her successful defense. Ruby is the first PhD from my lab. Special thanks to collaborators & examiners @INM7_ISN @AvramHolmes @mertrory @gordonneuro @AlexandSchaefer @zuoxinian @HelenJuanZhou @DanielMargulies @pretty_Jingwei @csabaorban

Tweet media one

5

5

64

@bttyeo

Thomas Yeo

3 years

I am going to quote this thread next time reviewers complain to us about our effect sizes.

@BrianNosek

Brian Nosek (@[email protected])

3 years

Sharpen your intuitions about plausibility of observed effect sizes. r > .60? Is that effect plausibly as large as the relationship between gender and height (.67) or nearness to the equator and temperature (.60)?

Tweet media one

10

263

780

0

2

63

@bttyeo

Thomas Yeo

4 years

Our work on autism heterogeneity is now in press at Congrats to @tangsy93 & @SunNanbo , who did most of the work!

@bttyeo

Thomas Yeo

5 years

1/4 New preprint () with @tangsy93 @SunNanbo @DoriFloris & Adriana Di Martino. Autism is well-known to be heterogeneous, but (categorical) subtypes miss out on continuous (dimensional) inter-individual variation. So we used a Bayesian model to ...

Tweet media one

4

60

120

0

19

64

@bttyeo

Thomas Yeo

4 years

Respiratory variation (RV) causes large fMRI signal fluctuations across the brain. What about heart rate variability (HRV)? One might think so, but HRV usually co-varies with RV, so how can we be sure? Check out @csabaorban poster #2333 at #OHBM2020 #OHBM2020Posters @OHBM

Tweet media one

2

21

63

@bttyeo

Thomas Yeo

7 years

static FC + one-lag FC (aka 1st order AR model) can explain sliding window correlations very well @HelenJuanZhou

Tweet media one

5

43

59

@bttyeo

Thomas Yeo

5 years

Paper is now out at @NeuroImage_EiC . Free version here:

@bttyeo

Thomas Yeo

5 years

1/6 Preprint led by @pretty_Jingwei @rubykong92 with @AvramHolmes @getian107 @mertrory @R_Liegeois @csabaorban Global signal regression (GSR) is highly controversial: it's great at removing global artifacts, but introduces negative correlations, etc

Tweet media one

Tweet media two

3

81

149

2

21

63

@bttyeo

Thomas Yeo

4 years

I have a thalamic parcellation from 2010 that is still unpublished. I have been sharing it with everyone who asks for it, so at least it’s not wasted.

@j_bertolotti

Jacopo Bertolotti

4 years

How old is your oldest manuscript you actually made some way into writing it, but never actually managed to submit (because of reasons)? Mine is getting >6 years old by now ☹️

40

6

59

9

5

63

@bttyeo

Thomas Yeo

5 years

I have quite a few collaborators whom I have emailed/skyped many times, but never met in person. It's such a weird experience to finally meet them in person... I feel like I know them very well, yet feel strange to, e.g., give them a hug. Someone should do a fMRI study on this.

3

0

62

@bttyeo

Thomas Yeo

4 years

Yes! Found a way to subtly work "person, woman, man, camera, tv" into my lecture tonight.

2

1

62

@bttyeo

Thomas Yeo

4 years

Movie pitch: Alien invasion but aliens managed to fund an alien invasion denial campaign so half the population refuses to recognize an alien invasion even when the aliens are raining fire on the west coast.

2

2

60

@bttyeo

Thomas Yeo

1 year

Our last publication in Neuroimage I guess I am resigning from BWAS? @smarek0502 @tervoclemmensb @ndosenbach @DrDamienFair Congrats to @Leon_Oo1 @JianzhongChen9 Thanks to co-authors @ZShaoshi @jingwei_ava_li @INM7_ISN @danilobzdok @AvramHolmes

@bttyeo

Thomas Yeo

1 year

Still no contact from headhunters but perhaps this awesome preprint led by @Leon_Oo1 & @JianzhongChen9 will convince them! Here, we examine the relationship between feature importance reliability & prediction accuracy. What is feature importance?

2

22

47

0

23

58

@bttyeo

Thomas Yeo

1 year

At this rate, a joke about my resignation is going to get more attention than my papers on Twitter. I guess I have to resign from something every time I publish a paper 🤣

@bttyeo

Thomas Yeo

1 year

TFW my resignation gets more attention than my papers on Twitter.

6

1

195

1

0

59

@bttyeo

Thomas Yeo

5 years

Thanks to @Jadecci & @pretty_Jingwei , the standalone library for converting between MNI coordinates and fsaverage vertices are now available:

@bttyeo

Thomas Yeo

5 years

We want to write a library of functions (), so users can input MNI coordinates & get fsaverage vertices & vice versa. How should we represent fsaverage vertices? NB: fsaverage has 163842 vertices; USS = user-specified surface, e.g. white, pial... @Jadecci

7

8

33

1

24

59

@bttyeo

Thomas Yeo

2 years

1/2 Don't forget to check out our epic 49-page reviewer response letter! An interesting issue pointed out by the reviewers was that our QC (while standard) led to a sample less representative than the full ABCD sample. Participants with higher quality data were more likely...

@angelatamtweets

Angela Tam | 譚

@angelatamtweets

2 years

2 years, 1 job change & 3 apartments after submission, my postdoc work with @bttyeo & co-author @JianzhongChen9 is finally published. We used fMRI functional connectivity to predict individual differences in cognition, personality & mental health in ABCD.

4

68

276

2

11

59

@bttyeo

Thomas Yeo

5 years

Preprint: with @Rajankashya @rubykong92 @pretty_Jingwei @HelenJuanZhou We found run-specific (state-specific) effects that were shared across HCP participants. By removing these common effects, we improve behavioral prediction accuracies.

Tweet card media

Individual-Specific fMRI-Subspaces Improve Functional Connectivity Prediction of Behavior

There is significant interest in using resting-state functional connectivity (RSFC) to predict human behavior. Good behavioral prediction should in theory require RSFC to be sufficiently distinct...

www.biorxiv.org

0

38

58

@bttyeo

Thomas Yeo

3 months

1/3 Backlog in announcing preprints, but here's goes! Preprint #1 by Pansheng Chen We previously proposed a meta-matching framework to translate predictive models from a single large dataset to predict new phenotypes in small data. Here we extend the ...

Tweet media one

1

27

58

@bttyeo

Thomas Yeo

2 years

Pre-trained model v1.0 from the UK Biobank (N = 36K) is now publicly available: Feedback is welcomed!

Tweet card media

GitHub - ThomasYeoLab/Meta_matching_models

Contribute to ThomasYeoLab/Meta_matching_models development by creating an account on GitHub.

@bttyeo

Thomas Yeo

4 years

1/5 Our new preprint uses meta-learning to boost functional connectivity based prediction of *NEW* phenotypes in small samples. We achieve 100%-400% improvements in many scenarios! @hetong93 @AvramHolmes @anlijuncn @danilobzdok @AvramHolmes @INM7_ISN

7

65

162

1

16

56

@bttyeo

Thomas Yeo

4 years

“deep learning ... can be a valuable tool for analyzing functional connectivity. However, Elastic Net probably performs better at present” — this resonates with our work showing kernel regression being competitive with DNN for FC-based prediction:

@richardfbetzel

Rick Betzel

@richardfbetzel

4 years

Wow -- this looks impressive | Optimising network modelling methods for fMRI |

2

40

90

3

15

54

@bttyeo

Thomas Yeo

4 years

Why do journals require the emails of all authors to be submitted but they only publish the corresponding author's email address? As an editor, I would like to invite ECRs to review, but finding their emails is pretty hard.

2

1

55