How can AI succeed in hospital care? On September 28, the AMPEL team will be at the Saxon Hospital Day to present the AMPEL-CDSS as a successful example. We are looking forward to stimulating conversations!

More information: https://khg-sachsen.de/kgs-engl/

#MedicalAI #AMPEL #Saxony #eHealth #CDSS

How can we enable local health systems to effectively govern medical AI?

Mark Sendak, Karandeep Singh, Suresh Bali, & I argue in a piece (out today in Nature Machine Intelligence) that a) this is a tough task; b) there some tasks that are going to be unavoidably local; and c) there are some policy ideas that can help enable collaborative governance.

Enjoy!

Free full version at https://rdcu.be/diYpU; feel free to share!

#AI #health #medicalAI #governance #lawfedi

When AI is used in medicine, things will go wrong & patients will get hurt.
(Just like medicine *without *AI! Hopefully less!)

Who *should* be liable?

Glenn Cohen & I have a surprisingly (we think) simple answer: hospitals. They're the last mile & can deal with contextual bias & in situ implementation. Lots of details! Default rules, switching! But that's the base.

Draft up at https://ssrn.com/abstract=4517740, forthcoming DePaul L Rev. Comments welcome!
#LawFedi #MedicalAI #Tort #AI #HealthLaw

I have created a "Medical AI" list of mastodon users to help colleagues get started more quickly #twitterMigration

You can add yourself here:
https://followlists.online/list/42dd6cc8-e347-d0b1-fcdc-72e9c254626c

#MedicalAI #ML #CDSS #CDS #eHealth #DigitalHealth #med #AI

Why OpenAI’s ChatGPT Is Such A Big Deal https://youtu.be/pOmpqdlVCoo @Youtubeより
I'm a Japanese pharmacist and want to be the capable brain, but I think that the both medical and pharmaceutical knowledge of OpenAI is insufficient.
However, if AI continues to evolve, there will be the possibility that AI takes over the jobs of both physicians and pharmacists.
#ChatGPT #chatgpt3 #AI #Pharmacist #ArtificialIntelligence #OpenAI #MedicalAI

I am still getting used to Mastodon, and just realized that without hashtags this is unsearchable!
#patent #patentfedi #biotech #AI #medicalAI #lawfedi #innovation

Out now “Levels of explicability for medical artificial intelligence: What do we normatively need and what can we technically reach?"
With Ursin, Lindner, Ropinski and Salloch on #informedconsent in #MedicalAI #bioethics https://link.springer.com/article/10.1007/s00481-023-00761-x

RT @NYSORA
🤖 NYSORA's latest YouTube video talks about the impressive trajectory of AI in #regionalanesthesia and #pocus. Are you curious to see what's next in the world of #medicalAI?

Watch our new video here: https://www.youtube.com/@nysoravideo

🚨 Paper alert 🚨

Our paper "RegionWise loss for biomedical image segmentation" (w/ @twitter@jussi_tohka) was recently published in Pattern Recognition journal. Link.

Here, we introduce a new stable way to optimize CNNs to account for pixel importance. What does this mean?

#MedicalAI #MedicalImaging #NewPaper

👇🧵

Good to see more people here.

If you are into #MedicalAI #MedicalImaging I recommend you to follow @fepegar - creator of the best Pytorch library for data augmentation in medical images.

A tardy #introduction

I study (mostly) how law shapes biomedical innovation, including #BigData & #MedicalAI, drug manufacturing, and associated topics.

I teach at #MichiganLaw #GoBlue

Mildly goofy, highly enthusiastic, extremely colorful, quasi-serious sometimes.

I love puns, #sciencefiction & #fantasy, good food, board games, #cooking, tasty spirits.

See you around!

If you are an expert in Reinforcement Learning and want to do research in this field, we are looking for PhD and Postdocs in a very impactful project named IntelliLung where we are developing AI-based approaches for assisting Clinicians in ICUs. details at: https://lnkd.in/eSVZWyjF #reinforcementlearning #medicalai #ai4good #artificialintelligence #phdresearch

I'm quite surprised to see so many papers fine-tuning ImageNet-pretrained models on medical data.

Aside from the major differences in the data, resizing images from, e.g., 2000x2000 to 224x224, to fit, say, a ResNet50, is a huge loss of information. And medical images are very similar within each dataset compared to ImageNet, i.e., their relevant information is likely to be on a few pixels.

#MedicalImaging #MedicalAI

Hi 👋

It's always a good time to explain what is medical image segmentation (MIS) and why it is so important (and cool). I will do this in plain English, and I will also highlight some active research areas. Keep reading if you want to learn more 👇

🧵 #MedicalAI #MedicalImaging

RT @pranavrajpurkar

Medical data used to train AI algorithms lack diversity. Most data come from a few hospitals in the US.

Here's an initiative to bring you diverse medical image datasets from across the globe: MAIDA.

https://www.rajpurkarlab.hms.harvard.edu/maida

#MedicalAI #MachineLearning #ComputerVision

So new to the app. Is there a niche group for #medtech or #MedicalDoctor #MedicalAI

I made a Medical AI a.gup.pe group, @medicalai

#MedicalAI

#introduction
Hi everyone, I'm so excited to learn about #machinelearning and #medicalai with all ya'll cool scientists.

I am a senior PhD student at the intersection of optics and ml. My first three years were devoted to non-linear microscopy and taking images of collagen. Then... I let the darkside takeover and I learned #python - thankfully I had a background in C++ and matlab.

Now my project uses a modified #stylegan to explore collagen features in different diseases.

Hi, #MedMastodon,

I'm a #PediatricOrthopedic #Surgeon
#artist and #medicalAI #Developer

I use #Art and #Painting via the #ArztForACause movement to generate awareness about #medical #diseases especially #cerebralpalsy and #limb #deformity

I recently tried four different implementations of Dice loss and the final performance was quite different. Implementation matters #MedicalAI