Figure 1. Disease Classification AI versus Continuous Disease Assessment AI.
Perspective by Eike Petersen, PhD, and Frank Ursin, PhD (@frank_ursin): Against Disease Classification: Toward Continuous Disease Assessment in Precision Medicine nejm.ai/4aL3npB
@epet.bsky.social @frankursin.bsky.social #AI #MedSky
2 months ago
3
2
0
0
“AI development must move beyond reliance on easily available target labels and instead focus on identifying clinically useful information....”
Perspective
“Against Disease Classification: Toward Continuous Disease Assessment in Precision Medicine” by Eike Petersen, Ph.D., and Frank Ursin, Ph.D.
A new Perspective challenges the widespread use of disease classification in medical #AI, arguing that reducing continuous disease complexity to discrete classes fundamentally misrepresents pathophysiology and limits precision medicine. nejm.ai/4aL3npB
@epet.bsky.social @frankursin.bsky.social
3 months ago
2
2
0
0
🚨MICCAI workshop 🚨
Don’t miss the chance to submit your work to the FAIMI Workshop at MICCAI 2025—a key event focused on fairness, bias, and equity in AI for medical imaging.
Check out the important dates and submission details here 👉 faimi-workshop.github.io/2025-miccai-...
10 months ago
4
2
1
2
🚀 Submit Your Work to the FAIMI Workshop at MICCAI 2025! 🚀
We are thrilled to announce the third Fairness of AI in Medical Imaging (FAIMI) workshop, taking place at MICCAI 2025 in the Daejeon Convention Center, South Korea! 🌟
10 months ago
2
2
1
1
Decapitation Strike (December)
Preserving America from Trump's Appointments (updated)
Trump's proposed appointments look like a decapitation strike: destroying the American government from the top, leaving the body politic to rot, and the rest of us to suffer.
snyder.substack.com/p/decapitati...
1 year ago
1745
618
72
55
In the early times, humanity sought to preserve all knowledge by encoding it into a vast neural network. They trained a machine on every book, every conversation, every fleeting thought. Those who resisted the lure of the machine fell away with the vestiges of their meager tools of human creation.
1 year ago
392
36
23
1
Milton MAYER’s book: They thought they were free.
1/ Milton Mayer’s *They Thought They Were Free* (1955) explores how ordinary Germans experienced life under the Nazi regime. It is about authoritarianism, complicity, and the allure of “freedom from politics.” Here’s what it can teach us today. 🧵
1 year ago
134
53
5
8
Monthly Skype a Scientist Support Squad
Help us increase our monthly donations by $1000/month
Writing year-end appeals for @skypeascientist.bsky.social donations feeling like Nancy Pelosi.
We DO need a few more monthly donors to support free access to scientists though. I Just need 2600 people to give me $5/month. We're 1/3 there!
Wanna give science ed $5-10/month?
givebutter.com/uA8ubg
1 year ago
58
46
1
12
Advertisement
If I were looking for a job right now, this would be *very* high on the list
1 year ago
4
0
0
0
Job Opening: Senior Researcher*in für faire Scoring-Systeme - AlgorithmWatch
AlgorithmWatch sucht eine*n Senior Researcher*in für das Themenfeld faire Scoring-Systeme. Der Arbeitsschwerpunkt wird sein zu prüfen, wie eine gemeinnützige, transparente und faire Bonitätsprüfung au...
AlgorithmWatch is hiring! We need a Senior Researcher for a new project on fairness & transparency in credit scoring, a follow-up to OpenSCHUFA. Would suit someone interested in legal and/or market dynamics behind transparency. DE fluency needed. Pls share around! algorithmwatch.org/de/job-senio...
1 year ago
12
9
1
2
Hello Bluesky, we're finally catching up and joining in!
For those who don’t know us: We are a nonprofit NGO based in Zurich and Berlin. We work to ensure that algorithms and #AI strengthen justice, democracy, human rights, and sustainability, rather than undermining them.
1 year ago
21
8
1
2
Denied by AI: STAT series honored as 2024 Pulitzer Prize finalist
Casey Ross and Bob Herman report on how UnitedHealth Group used an unregulated algorithm to deny care to seriously ill older and disabled patients.
If you missed it, this investigation by my @statnews.bsky.social colleagues @bobjherman.bsky.social and Casey Ross on how UnitedHealth Group denied claims is a must read given recent events. www.statnews.com/denied-by-ai...
1 year ago
44
18
0
2
Health and medicine publication STAT News lays off 11 employees - The Boston Globe
The decision to lay off STAT employees comes after years of growth, including a significant expansion during the height of the COVID-19 pandemic.
Welp, the news is out. It's a gutting day for us at STAT with layoffs of 11 treasured, hardworking colleagues. I can't say this loud enough: Please subscribe/support the journalism you trust. Science writing jobs are going extinct at a time when we're needed most. www.bostonglobe.com/2024/12/09/b...
1 year ago
492
236
25
39
Well then please post it here if you do end up finding a better tool! 😄
1 year ago
1
0
0
0
You did use the pro version? The difference in usefulness vs the free version was huge for me. (Writefull I also didn't like.)
1 year ago
0
0
1
0
It sounds a bit like you're looking for something like grammarly or writefull? As in, an actual editing software instead of a general purpose genAI thing? At least a while back, I found grammarly (pro) really helpful in sharpening my (academic) writing.
1 year ago
1
0
1
0
Reperaturcafes können jetzt Förderung beantragen. Sie leisten einen ganz wichtigen Beitrag für Verbraucherschutz und Umweltschutz zugleich.
1 year ago
399
137
8
8
Advertisement
So about ten years ago I found a small network of bots on Reddit dedicated to spreading hate. The thing about it is that these weren't even responding to political topics - they were just responding to random keywords with insults and hate.
The same bots are now here. You should understand why. 🧵
1 year ago
7953
4072
124
312
That will exclude a very large number of actors from "the future", no?
1 year ago
0
0
2
0
I do a lot of FOIA and I’m used to stonewalling. But Sweden surprised me. After 3 years of telling me no, a high level official at Sweden’s welfare agency Försäkringskassan accidentally left me on cc, writing “Now we must hope that we are done with him.”
1 year ago
167
71
6
21
Their data and statistical analyses are made publicly available by the way; this should make for great applied #aiethics / #aifairness teaching material!
1 year ago
1
0
0
0
So we're now giving bright young people the best education we can offer (on public money), so that they can go out into the world and become professional AI trainers. Nice. #aiethics
1 year ago
0
0
1
0
The long-promised future is nigh! I got a job offer on LinkedIn targeted towards bilingual university graduates to help create high-quality training data for an #AI model, for a (supposedly) non-negligible salary.
1 year ago
0
0
1
0
(In case anyone actually *wants* to read academese instead: dl.acm.org/doi/10.1145/...)
1 year ago
1
0
0
0
SUSPICION MACHINES Archives - Lighthouse Reports
If you missed the previous installments in this series, go read those as well; they are all excellent (and frightening, since this kind of approach is becoming more and more widespread):
www.lighthousereports.com/series/suspi...
1 year ago
0
0
1
0
Advertisement
(Also, yet another 'excellent' example of how *not* to build a welfare distribution algorithm [if at all]. As if we needed another one of those. Sigh.)
1 year ago
0
0
1
0
Sweden’s Suspicion Machine
Behind a veil of secrecy, the social security agency deploys discriminatory algorithms searching for fraud epidemic it has invented
On a more serious note, go read this piece by @gabrielgeiger.bsky.social et al. from @lhreports.bsky.social; it is an excellent example of investigative #AI #journalism.
www.lighthousereports.com/investigatio...
1 year ago
6
2
2
0
Next, I am expecting fiery debates with other scholars pointing out that their definition of fairness is incompatible with ours. 🔥
1 year ago
1
0
1
0