Thomas Nichols

Alcohol use and risk of dementia in diverse populations: evidence from cohort, case–control and Mendelian randomisation approaches Objectives To investigate the relationship between alcohol consumption and dementia. Design Prospective cohort and case–control analyses combined with linear and non-linear Mendelian randomisation. ...

Happy to be a part of Anya Topiwala's new work, using Mendelian Randomisation with @ukbiobank.bsky.social & Million Veteran Program, finding that even modest alcohol consumption increases risk of dementia. ebm.bmj.com/content/earl...

𝐂𝐚𝐥𝐥 𝐟𝐨𝐫 𝐏𝐚𝐩𝐞𝐫𝐬 – 𝐒𝐩𝐚𝐭𝐢𝐚𝐥 𝐒𝐭𝐚𝐭𝐢𝐬𝐭𝐢𝐜𝐬 𝐢𝐧 𝐈𝐦𝐚𝐠𝐢𝐧𝐠

Special issue of 𝘚𝘵𝘢𝘵𝘪𝘴𝘵𝘪𝘤𝘴 𝘢𝘯𝘥 𝘋𝘢𝘵𝘢 𝘚𝘤𝘪𝘦𝘯𝘤𝘦 𝘪𝘯 𝘐𝘮𝘢𝘨𝘪𝘯𝘨, supported by @amstatnews.bsky.social ASA Sections on Statistics and the Environment (ENVR) and Statistics in Epidemiology (SIE)

Submission details:
tinyurl.com/47enwhwz

AI-driven reclassification of multiple sclerosis progression - Nature Medicine Using large cohorts from published clinical trials involving more than 8,000 patients with multiple sclerosis, a probabilistic machine learning model reconstructs the transition probabilities from data-derived diseases statuses, showing patterns that suggest how progression to severe stages occur and potential inversion of the process.

All made possible by an innovative Factor Analysis + Hidden Markov Model. See paper & supp for details

Ganjgahi et al. ‘AI-Driven Reclassification of Multiple Sclerosis Progression’. Nature Medicine, Nature Publishing Group, 20 August 2025, 1–11. doi.org/10.1038/s415....

Well known diagnostic labels (RRMS, SPMS, PPMS) do not explain these states, showing their limitations. This is an important contribution to developing more fully data-driven perspectives on MS disease heterogeneity.

This is a key output of the Oxford Big Data Institute-Novartis Collaboration for AI in Medicine. The paper represents a huge, collaborative effort, with replication done on data from another pharma company (Roche) and another large dataset (MS PATHS).

www.bdi.ox.ac.uk/research/oxf...

Epic work led by Habib Ganjgahi is out in Nature Medicine... data-driven clustering of MS finds 3 distinct states, early/mild, active, advanced, w/ transition to the advanced state only occuring through the active state... replicated twice!

doi.org/10.1038/s415...

This is a gem… really enjoyed reading about stats theory legend Gauss trudging across the countryside surveying.

Really nice study, and extends some of the ideas developed in this paper pubmed.ncbi.nlm.nih.gov/32673043/

For me, this work is a classic @ohbmofficial.bsky.social story: In 2023 I wasn't working with @bttyeo.bsky.social but I overheard him at his poster pointing to some accuracy curves saying "I don't why they have this particular shape". That kicked off the collab that led to these results.

1/11 Excited to share our @Naturestudy led by @leonooi.bsky.social @csabaorban.bsky.social @shaoshiz.bsky.social

AI performance is known to scale with logarithm of sample size (Kaplan 2020), but in many domains, sample size can be # participants or # measurements...

doi.org/10.1038/s415...

Wasn't aware of tinyplot... an overlay package to make base R graphics easier to use. While I use ggplot2 for some tasks there are many situations where I prefer base R plotting, and this tool will be super helpful in those seetings.

Artificial intelligence for electronic health records For our next talk, in Machine Learning, we will be hearing from Dr Shishir Rao, Senior Research Scientist, Deep Medicine research group, NDWRH, University of Oxford. We’re delighted to host Shishir in...

Looking forward to presenting our team's research on AI for EHR at The Big Data Institute at Oxford on July 10th at 2 pm! Thanks to Sumeeta and Prof @nichols.bsky.social for organising.

Link: talks.ox.ac.uk/talks/id/590...

Excited to share our latest preprint: "Neurosynth Compose: A Web-Based Platform for Flexible and Reproducible Neuroimaging Meta-Analysis"

Here, we cover the why, what, and how of the next generation Neurosynth ecosystem for reproducible neuroimaging meta-analysis.

osf.io/preprints/os...

Almost recovered from #OHBM2025... has anyone else found the E-posters nearly imposible to find. They are here, discoverable through the app and the browser-link-out button

ww6.aievolution.com/hbm2501/inde...

Public service announcement: Don’t forget to upload your #OHBM2025 poster PDF, at ww6.aievolution.com/hbm2501/

The beta release of SPM-Python is out now! Amazing work by Johan Medrano @johmedr.bsky.social , Yael Balbastre, Yulia Bezsudnova @ybezs.bsky.social and other members of their team. A new era for SPM! #OHBM2025

Interested in practical tips for assessing collinearity and why common fixes might not help? Come to the #OHBM2025 oral session Friday at 11:30AM (P2)! I’ll share simulation results and @demidenko.bsky.social will present real data. Also, check out my poster Thursday (1051)!

Is anyone else at #OHBM2024 educational courses maddened by the lack of time per-course schedule? In the app, the individual talks are listed *alphabetically* by talk title, all with the same start time. Same thing for symposia...

FYI, this link comes from Poster Presenter's page
humanbrainmapping.org/i4a/pages/in...

Useful #OHBM2024 deep link - Table listing all posters with stand-by times & categories

gmpartners365-my.sharepoint.com/:x:/g/person...

FYI, linked from Poster Presenters page
humanbrainmapping.org/i4a/pages/in...

Preserve the Open Science Room at the Organization for Human Brain Mapping Annual Meeting Can you spare a minute to help this campaign?

At #OHBM2025 the Open Science Room has been squeezed and next year is at risk of being dramatically reduced/cut. If you like the OSR please consider signing/sharing this petition! chng.it/yjs4Bnyf9s
@ohbmofficial.bsky.social @ohbmossig.bsky.social

Home What can we do with functional MRI data? For many years, fMRI has been used to discover population-level patterns of brain function, organization and connectivity, and to understand differences in t...

Looking forward to #OHBM2025 in Brisbane next week! My lab is recruiting a postdoc & neuroimaging analyst/developer to support our NIH funded work in fMRI-based Alzheimer's biomarker development (NIA R01AG083919). Email or DM if you want to meet up in Brisbane! www.statmindlab.com/join-us

Happy to see Julio Peraza's lastest posted, "NiCLIP", using a CLIP model to improve 'reverse inference' prediction of task fMRI results based on peak coordinate data www.biorxiv.org/content/10.1... w/ @angielaird.bsky.social @neurozorro.bsky.social James Kent & JB Poline

Happy to see my first first-author paper in Imaging Neuroscience (along with @fmrib-karla.bsky.social and @nichols.bsky.social):

Characterising ongoing brain aging and baseline effects from cross-sectional data

doi.org/10.1162/IMAG...

Registration Open! FSL Course 2025 will be held fully online, 15th – 26th September. Details & registration info here: open.win.ox.ac.uk/pages/fslcou...
The course covers lectures & hands-on practicals on structural, functional, diffusion and resting state brain image analysis.

An infographic providing an overview of the CODE roadmap for sharing research software and code: Open Document Execute Collaborate.

🤔 You’d like to share your research code but not sure how to…?

🔓*Very* excited to finally share this work led by @nicolasrougier.bsky.social in which we propose a practical set of guidelines for opening code.

inria.hal.science/hal-04930405...

Looks like a very useful tool...

Voets et al. outline best practice recommendations from the OHBM Working Group on clinical language fMRI mapping: doi.org/10.52294/001...

@cyrilrpernet.bsky.social @lindadouw.bsky.social @nichols.bsky.social @fmri-today.bsky.social @mallarchak.bsky.social @ohbmofficial.bsky.social

Kudo's to the statsupai.org/team.html for bringing this all together