Johannes Lotz

Had fun presenting with Carol Geppert at #ecdp2025.

Our poster “Image registration accelerates (pathology) diagnostics” shows that #imageregistration saves one minute (33%) per case in our pilot study!

Could not make it? Have a look the poster s.x91.de/ecdpposter and video s.x91.de/tnvideo

I have chills

RCV is a structure to build durable coalitions

Think how much healthier this is for the local party, for the city, for the actual final outcome

Sorkin-sentimental fantasy to imagine a politics like this

Live updates: Conclave elects Cardinal Robert Prevost to be Pope | CNN Cardinal Robert Prevost of the United States has been selected as the next pope, the first American pontiff in history. Follow here for the latest.

Adding yet another job to the “what can I do with a bachelor’s in math?” list 💪 #mathSky #iTeachMath

Man muss in der Debatte um die AfD daran erinnern, dass das Grundgesetz nicht irgendeine liberale Verfassung ist. Es ist die Antithese zum NS-Staat. Die Bundesrepublik ist politisch keine ergebnisoffene Veranstaltung. Grundrechte, Demokratie- und Rechtsstaatsprinzip sind dauerhaft festgelegt.

NEW 🧵

The number of people travelling from Europe to the US in recent weeks has plummeted by as much as 35%, as travellers have cancelled plans in response to Trump’s policies and rhetoric, and horror stories from the border.

Story: www.ft.com/content/6dc1...

Do foundation models in medical imaging see everyone fairly?🤔

Excited to share our new Science Advances paper uncovering & auditing demographic biases of expert-level VLMs, and comparing to board-certified radiologists🧑‍⚕️

📄science.org/doi/10.1126/sciadv.adq0305
💻github.com/YyzHarry/vlm-fairness
(1/)

French scientist denied US entry after phone messages critical of Trump found France’s research minister said the scientist was traveling to Houston for a conference when his phone was searched

Breaking: US authorities thwart an attempt to smuggle sanity into the country.

www.theguardian.com/us-news/2025...

Wie war das mit den Tassen? Völlig absurd.

Schön auch dazu:

BM Bastian Langbehn (fraktionslos) Änderungsantrag zu - AM Andreas Zander (CDU)

Beschlussvorschlag
In die Liste der Unterstützer:innen werden
Die Unklar.Bar
Die PARTEI Lübeck
mit aufgenommen.

1/2

Politik Informationssystem Lübeck - Rathaus

In Lübeck stellt die CDU ebenfalls so eine Anfrage und listet u.a. Schüler:innenvertretungen einiger Lübecker Schulen auf. www.luebeck.de/de/rathaus/p...

Amerika diskutiert das Ende der Sommerzeit und nicht mal Trump gelingt es (weil die einen Sommer-, die anderen Winterzeit wollen). Aber: Hawaii und Arizona sind schon weiter. Lasst es uns auch so machen: Wenn Schleswig-Holstein bei Winterzeit bleibt, dauert die Zugfahrt Lübeck-Hamburg nur -16 min.

Cartoon showing a woman looking in to a microscope saying "Whoaah!!!" The caption states "The best part of Science is knowing, for a moment, something that nobody else in the world knows"

A reminder in these hard times. #sciencestrong

If pubmed is inaccessible at the moment, try @europepmc.org

Eine simple Strichmännchen-Zeichnung: Links steht eine Figur mit beiden Händen an den Kopf gefasst, die laut “Aaaaaaaah!!!” schreit. Ein Zickzack-Blitz symbolisiert den Schrei. Rechts stehen zwei andere Strichmännchen. Das eine sagt zum anderen: “Er spricht aus, was viele denken.” Die Zeichnung ist mit “@kriegundfreitag” signiert.

Hier eine erste Analyse der Situation.

Da haben die Kolleg*innen eine gute Seite eins hinbekommen. Ganz kleine Anfrage an die Union: Noch alle Tassen im Schrank? Nailed it.

Really disappointed how slowly Europe (especially Germany) is moving these days
Obviously, to compete with US/China on any relevant topic (AI / economy / defense), we need more European integration, joing funding, joint efforts - and much much faster
more resources: eu-acc.com @euacc.bsky.social

Approximately 10% of the NIH staff were fired yesterday. Barely anyone outside of the scientific community noticed, and I'd wager a sizeable number of NIH funded PIs have no idea either.

Improvements in LM (top) and EM (bottom) of our micro-sam model (finetuned) compared to the default SAM model.

After a long journey, Segment Anything for Microscopy is now published in Nature Methods! We significantly improve SAM for interactive and automatic segmentation in light and electron microscopy and build a user-friendly tool.
www.nature.com/articles/s41...

To be clear: was hilft gegen Influenza?
1. Impfung

(lange nichts)

(Sehr lange nichts)

Dann irgendwann Maske tragen, desinfizieren und Abstand halten.
Aber -

Trump hits NIH with ‘devastating’ freezes on meetings, travel, communications, and hiring Researchers facing

Trump hits NIH with ‘devastating’ freezes on meetings, travel, communications, and hiring www.science.org/content/arti...

Big shoutout to @tillnitus.bsky.social (who did the lion’s share of the work 🦁), to our co-authors Raphael Schäfer, Henning Höfener, Friedrich Feuerhake, Dorit Merhof, and Fabian Kießling and to the whole team at FraunhoferMEVIS! Special thanks to the reviewers for their constructive feedback!
🧵5/5

GitHub - FraunhoferMEVIS/MedicalMultitaskModeling: Training foundational medical imaging models using multi-task learning. Training foundational medical imaging models using multi-task learning. - FraunhoferMEVIS/MedicalMultitaskModeling

You can already find the base model on GitHub (github.com/FraunhoferME...), the pathology-specific model will follow shortly.

🧵4/5

... CTransPath (by Wang et al., doi.org/10.1016/j.me...) and additionally, we observed similar results when comparing against the larger UNI model (by Chen et al., doi.org/10.1038/s415..., see the appendix for more details).

🧵3/5

“Tissue Concepts” is a resource-efficient foundation model for pathology. We trained with only 6% of the data compared to other (self-supervised) models, and it maintained great performance.

It outperforms self-supervised learning when the model is of similar size (we compared to ...
🧵2/5

Our paper on Tissue Concepts is finally out in JCompBiomed 🥳: Supervised pre-training of foundation models needs fewer data and less compute/energy to analyze pathology whole-slide images.

How? Read more here: doi.org/10.1016/j.co...

🧵1/5

I missed this one last month. New AI tool predicts gene expression from pathology slide images:

Digital profiling of gene expression from histology images with linearized attention https://www.nature.com/articles/s41467-024-54182-5 🧬🖥️🧪 https://github.com/gevaertlab/sequoia-pub

Eleven quick tips for properly handling tabular data https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1012604 🧪 #Rstats

Zugverspätungen: Der ehrliche Fahrplan: Mit welchem Zug Sie rechtzeitig ankommen Sie müssen pünktlich zu einem Termin ankommen? Unser Tool zeigt, ob das mit Ihrer Verbindung realistisch ist. Wenn nicht, haben wir eine zuverlässige Alternative für Sie.

Wir sind an dem Punkt, an dem mit Hilfe von statistischen Daten und Algorithmen die Wahrscheinlichkeit von Verspätungen bei der Bahn recht zuverlässig berechenbar ist.

Paywall > Freebie/Geschenklink

Gleason 6: should we keep calling it cancer?
Eva Compérat

Yes‼️ It’s cancer
✨People may die from GS 6
✨Up to 80% heterogeneous
✨We can not predict those who will progress
‼️Patient will be much more willing to be followed
@oncoalert.bsky.social
@drnataliagandur.bsky.social
@yukselurun.bsky.social

Hi, I would be happy to be added!