Nicola Romanò @nicolaromano@qoto.org
Senior lecturer at the Zhejiang-Edinburgh Joint Institute (ZJE) and Edinburgh University.
Undergraduate Programme Director, Biomedical Informatics at ZJE.
I teach #imageanalysis & #dataanalysis with #RStats & #python. I study #heterogeneity in #pituitary (and other) cells.
I'm also very interested in #reproducibility and #openscience.
Joined Oct 2022
Nicola Romanò
boosted
As a Blind person i never thought i would be on social media savoring photos. But the communal Mastodon alt text game is so strong that sweet, poetic or silly descriptions abound on my timeline. Thanks to legions of people who take time to write a meaningful description of the ephemera they post, i learn so much about insects, plants, buildings, memes — all dispatches from a dimension of the world that i otherwise wouldn't experience. If you're wondering whether anybody reads these things: YES.
@cybercow Hype and marketing, coupled with a general misunderstanding of how ML and AI work.
"Powered by AI" sells products faster, whether it's true or not that AI has something to do with the product, and whether it actually improved it in any way
"Arbaugh initially faced issues after his surgery when the tiny wires of his implant retracted, resulting in a sharp reduction in the electrodes that could measure brain signals. Reuters has reported *Neuralink was aware of this issue from its animal trials*. [...] Musk also said he has spoken with Donald Trump, who he has endorsed in the US presidential race, about forming a commission aimed at improving “government efficiency” through *reduced business regulation*, and would be willing to participate. Musk said that in his view US regulations hamper innovation."
Of course, ethical and safety considerations as usual are not in Elon Musk's book; nor is openness.
Brain computer interfaces are the last area where you want reduced regulation...
Nicola Romanò
boosted
Did you forget to write some slides? do you want something quick and easy? try using my new quarto revealjs plugin: loud
All text on slides are automatically centered and zoomed in to fill the screen
Nicola Romanò
boosted
"Our framework suggests that after accounting for heterogeneity, the probability that the tested hypothesis is true for the average population, design, and analysis path can be much lower than implied by nominal error rates of statistically significant individual studies."
#MetaSci #Methodology 🧪
Heterogeneity in effect size e...
@elduvelle well, it's unknown to me, as it's a small journal focussing on something I generally don't look at. However looking at the website it seems legit.
I've had emails in the past from really random, probably predatory journals asking for review, though (say an engineering journal asking for a review given my expertise in building cars or something :/ )
How do you folks deal with random invitations to #review for #journals you never heard of?
I usually just delete the email and not engage as usually these are random topics I have zero expertise in... today, however, I received a paper about something (kind of) related to some work I've done in the past. Reading the abstract I can already see very major issues with it so... should I put time and effort to prevent obviously bad work to be published? Or is this a battle not worth fighting?
@MCDuncanLab yeah I was a bit baffled by the stash of totes as well, although I did use to have a stash of plastic bags under the sink... However I'm fascinated by the marketing that goes around this.
@CrisLuengo right, that makes sense, thanks for clarifying!
Interesting article on "reusable" products
"An entire generation has been socialized to participate in environmentalism by way of consumerism"
https://www.theguardian.com/environment/article/2024/jul/30/reusable-tote-bags
@CrisLuengo Thanks for raising that point. I have been thinking about this and I am wondering, though, wouldn't the same reasoning work for say, Sobel? Unless you threshold the gradient magnitude you are not really extracting edges by convoluting with a Sobel kernel.
Similarly, while the LoG is not per se an edge detector, finding its zero-crossings allows you to detect edges (indeed, you can also use LoG as a blob detector by finding its extrema).
So, is it just a question of semantics (LoG as a filter used in edge detection vs LoG as an edge detector), or is there more?
I am curious, as there is definitely a lot of literature saying that LoG is an edge detector.
New Python Shiny app for my image analysis course deployed!
This will accompany the lecture on edge detection methods.
https://apps.nicolaromano.net/EdgeDetection/
Source code is here, feel free to reuse!
https://github.com/nicolaromano/BIA4/tree/main/Apps/EdgeDetection
Nicola Romanò
boosted
@LianaBrooks excellent points! May I add:
Privilege comes in many shapes and sizes, it’s not all-or-nothing, so you can have various types of privilege without having every possible advantage in life. You can have struggles & suffer injustice and still have privilege in some form. You can have privilege in one area and lack of privilege in another.
"Freshening up" the material for my #biomedical #image #analysis course restarting in September.
A lot of my students will be essentially #Python novices... so here is some short Python starting/refresher material! Everything's CC-BY-4.0 so feel free to reuse!
https://github.com/nicolaromano/BIA4/tree/main/Workshops/00%20-%20Preliminary%20material
@tangming2005 Absolutely agree with this. I've recently scrapped a (fairly complex and difficult to interpret) NN pipeline for transcriptomics analysis for a much simpler XGBoost pipeline and performance is the same or higher in all cases.
Also, interpretability is much higher.
Nicola Romanò
boosted
Looking for literature for review paper:
Is there quantitative data that beauty/aesthetics in charts help user engagement/effectiveness/memory?
Nicola Romanò
boosted
Great new blogpost by Eiko Fried on “zombie theories” in science, research waste, and moving forward. #OpenScience #academia
https://eiko-fried.com/zombie-theories-why-so-many-false-ideas-stick-around/
Nicola Romanò
boosted
Everybody's Free (To Write Websites)
Enbies and gentlefolk of the class of '24:
Write websites. If I could offer you only one tip for the future, coding would be it. The long term benefits of coding websites remains unproved by scientists, however the rest of my advice has a basis in the joy of the indie web community's experiences. I will dispense this advice now:
Nicola Romanò
boosted
I can't believe that in 2024 we still have to explain so bluntly that "Academic journals are a lucrative scam". Yet here we are ...
Kudos to the Editorial Board of Philosophy & Public Affairs for resigning en masse and thank you to Arash Abizadeh for explaining why and what's next in the Guardian:
https://www.theguardian.com/commentisfree/article/2024/jul/16/academic-journal-publishers-universities-price-subscriptions
Nicola Romanò
boosted
"'With the R515 driver, #NVIDIA released a set of #Linux GPU #kernel modules in May 2022 as open source with dual GPL and MIT licensing. […]
Two years on, we’ve achieved equivalent or better application performance with our open-source GPU kernel modules […]
We’re now at a point where transitioning fully to the open-source GPU kernel modules is the right move, and we’re making that change in the upcoming R560 driver release. […]'"
https://developer.nvidia.com/blog/nvidia-transitions-fully-towards-open-source-gpu-kernel-modules/
Senior lecturer at the Zhejiang-Edinburgh Joint Institute (ZJE) and Edinburgh University.
Undergraduate Programme Director, Biomedical Informatics at ZJE.
I teach #imageanalysis & #dataanalysis with #RStats & #python. I study #heterogeneity in #pituitary (and other) cells.
I'm also very interested in #reproducibility and #openscience.
Joined Oct 2022
