rimmon @rimmon@qoto.org

A group at Johns Hopkins has created a scrollable, interactive map of the entire universe, from here to the cosmic microwave background.
Extraordinary discoveries at your fingertips for free, unimaginable when I was a kid. https://mapoftheuniverse.net/ #astronomy #space #exploration

Are those of you here in the #Fediverse aware of the amazing resource that is https://mynoise.net?

No?
Let me introduce you.

It has a medieval village #background #noise environment with toggles you can use to increase or decrease the blacksmith shop and market and horse clops (among many others).

Oh, you want a coffee shop? Sure thing.
Rain? Thunderstorm? Gotcha.
Generic white noise?
RPG thrills? Of course.
Tame that #tinnitus? Yes!

Check it out!

#soundscape #WorkSounds

Right now everyone is still engaged in making connections and introductions, but as @PessoaBrain mentioned, we should think about ways to stimulate good #neuroscience conversations (especially sans timeline algorithm).

I don't think there is any magic formula, so I'd encourage everyone to just post what fascinates them: stray thoughts, quotes from papers and books, diagrams, contentious topics, and perhaps my favorite: 'dumb' questions. 🤓

[...]

Academic jargon, translated

#academicmastodon

“You don’t look disabled”

#introduction

I'm Tyler Morgan (she/her), a postdoc at #sfim / #nimh / #nih working on #fMRI / #layerfMRI / #7TMRI / #mrirecon methods for #neuroscience projects.

I earned my PhD in Neuroscience & Psychology from the University of Glasgow (#UofG #PsychNeuro). That work focused on predictive and contextual information in cortical feedback.

NeuroImage

A large and rich EEG dataset for modeling human visual object recognition
https://www.sciencedirect.com/science/article/pii/S1053811922008758?via%3Dihub

Here's the workflow I had in mind.
1. Data Ingestion
2. ETL vs ELT
3. Data Lakes & Warehouses
4. Organizing ML models
5. Packaging
6. CI/CD
7. Serving
8. Testing
9. Orchestration & Monitoring
i. K8s
ii. Serverless
10. Event Processing
11. Data Drift

What would you do differently? Any resource recommendations?

I will never understand why people are so averse to diversity.

I like seeing trans people speaking to each other with idioms I don't get, cute girls speaking a language I can't recognize, or gay dudes priming and posing in random pics.

Seeing shit that is not for me makes me feel like I'm part of a much bigger world that still has space for me to learn and grow.

That's a big place of comfort for me. The bottomless nature of the human experience means we all have space to be fully who we are.

That's pretty cool.

Holy moly this thingy is sick!

... and this cheat sheet on curve fitting, by XKCD (on Masto?)...
https://m.xkcd.com/2048/

Here's a cool simulation to explore the way sampling error increases with measure complexity (from single mean, to mean difference, to 2-way interaction, etc.):

* Video: https://www.youtube.com/watch?v=dVAuWiOq4Qc
* Source code: https://osf.io/a8nxg

I made this as part of a talk on Statistical Thinking I gave at #SFN2022.

The simulation uses the amazing switchboard package by
@LajeunesseLab

#Statistics #Reproducibility #Science

These brains aren't real!

Very happy to share that our paper "A deep generative prior for high-resolution isotropic MR head slices" has been accepted to #SPIE Medical Imaging 2023!

We train StyleGAN3 on a carefully curated high-res #mri neuroimaging dataset and analyze the network manifold. Presentation will be in San Diego Feb 2023.

#phd #science #machinelearning #generative

Super busy here with preparing abstracts for today's #ISMRM conference deadline. Lots of nice results from highly interdisciplinary and collaborative projects; some foundational #MRI #physics and image reconstruction, #histology, #biophysical modeling, #qMRI, #hMRI, #diffusion MRI and layer-specific #fMRI.

https://www.ismrm.org/

The International Brain Laboratory is on Mastodon! We are 22 laboratories comprising experimental & theoretical neuroscientists. We collaborate to understand brainwide circuits for complex behavior. Looking forward to posting here! #neuroscience

Reminder that my deep learning course at the University of Geneva is entirely available on-line. 1000+ slides, ~20h of screen-casts.

Full of examples in PyTorch

https://fleuret.org/dlc/

How you’re born (vaginal birth vs c-section) influences your #microbiome which influences your response to #vaccination.

Relative abundances of vaginal birth-associated Bifidobacterium and Escherichia coli in the first weeks of life are positively associated with anti-pneumococcal and anti-meningococcal antibody responses.

https://www.nature.com/articles/s41467-022-34155-2

Those interested in scientific visualization (#scivis, #sciviz, #datavis, #dataviz) may enjoy the videos in the #HHMIJanelia Scientific Visualization Interest Group playlist:

https://www.youtube.com/playlist?list=PLfBkfxqisR_dU9RyIeX0I1WbQee86ZTss

These videos include talks from the creators of systems like Agave, ANARI, Datoviz, #napari, Neuroglancer, PyGfx, VizPy, and VMD. Thanks to @billkatz for organizing the series.

Today officially marks 8 billion humans on Earth. Meanwhile humanity has wiped out 70% of wild animal populations since 1970. We don't live outside of nature.