CardioTechx @CardioTechx

CardioTechx by @SaharSamimii & @DonnchadhOSull | Keep up-to-date with rapidly evolving Cardiology, Technology & Artificial Intelligence Research | Our own views Joined December 2024

Tweets

26
Followers

80
Following

95
Likes

44

Nikhil Krishnan @nikillinit

a month ago

How should we compare AI mistakes vs human mistakes in healthcare? Doctors make mistakes all the time, but it’s not like we remove the medical license of a doc if they make a few mistakes. The bar for AI is way higher - it has to be nearly perfect and tested in every possible…

18 9 58 10K 19

CardioTechx @CardioTechx

2 months ago

EchoNext, a DL ECG model trained on 1.2M ECG–echo pairs, prospectively identified previously undiagnosed SHD (PPV>50%) and exceeded cardiologist performance across 11 hospitals. Public weights + 100k-labeled ECG set released. doi.org/10.1038/s41586… #AIinMedicine #CardioTwitter…

0 4 12 2K 2

Download Video

Milos Vukadinovic @milos_ai

2 months ago

🚨Releasing 200k cardiac MRI images! 🚨 Open source cMRI datasets contain only a few hundred images, so we used GANcMRI to create 1000x bigger fully synthetic dataset. We hope that GANcMRI dataset will boost the development of AI models for cardiology.

3 4 17 3K 8

David Ouyang, MD @David_Ouyang

3 months ago

Excited to see this comparison of amyloid detection algorithms in @JACCJournals #advances, lead by @FarazAhmadMD. AI algorithms on #echofirst outperform EHR and rules-based approaches. Similar AUC but different calibration. @InVision_AI more specific, @ultromics more sensitive.…

0 10 34 2K 7

Download Image

Ultromics @ultromics

3 months ago

📢 New #EHJ study! Our AI screening tool for echo — EchoGo® Amyloidosis — detects cardiac amyloidosis, a life-threatening condition, with high accuracy. Developed with @MayoClinic @UChicago, tested with 18 global sites. See paper & release: ultromics.com/press-releases… #CardioTwitter

1 8 20 5K 2

Download Image

JACC Journals @JACCJournals

3 months ago

Pitfalls and promise in #AI/DL #EchoFirst: how to best evaluate the evaluation metrics to address bias? New #JACCIMG study evaluates 3 common challenges in EF estimation to mitigate these challenges for new AI/DL frameworks in #cvImaging. jacc.org/doi/10.1016/j.… #DeepLearning

0 7 24 4K 4

Download Image

CardiovascularCorner @TrackYourHeart

3 months ago

Open it. Zoom it. Save it. Share it. This is your go-to guide for standard adult TTE views.

2 85 346 40K 338

Download Image

CardioTechx @CardioTechx

3 months ago

🚨 Introducing ECGFounder, a foundation model trained on over 10 million annotated ECGs, achieving expert-level performance across 150 diagnoses. Validated across diverse datasets, it generalizes well, even with single-lead inputs. @NEJM_AI Link: ai.nejm.org/doi/abs/10.105……

0 10 27 2K 5

Download Video

CardioTechx @CardioTechx

4 months ago

🧠💓 Deep learning meets congenital heart disease. This AI-ECG model predicted future LV dysfunction months before echo. You need to see this. #AIinMedicine #Cardiology #MedEd @jmayour #CardioTwitter #CardioTech #ACHD #CHD #EchoFirst #HeartFailure2025 #HeartHealth…

0 3 10 2K 0

Download Video

Google @Google

4 months ago

Meet infinite possibilities with Flow and Veo 3. 🎬 ✨Catch up on all the latest Google AI announcements and demos from #GoogleIO → ai.google

0 384 4K 29.1M 1K

Arya Aminorroaya @AryaAminorroaya

5 months ago

After our recent 📖 featuring ADAPT-HEART, an AI-ECG model to detect structural 🫀 disease from 1-lead ECGs, thrilled to share another application of ⌚️/📱-based AI-ECG models to predict incident HF using 1-lead ECGs! Stay tuned for more from us @cards_lab @YaleCardiology!

CarDS Lab @cards_lab

5 months ago

1 9 27 5K 2

1 8 41 5K 1

Bardia Khosravi, MD @BrdKhsrv

5 months ago

Humbled that @NajvaApp has hit 1000+ users! I've dictated 100K+ words with it - it's become an inseparable part of my workflow. Voice-to-text + LLM is the ultimate productivity tool. Try it yourself (always free) and feel the magic! 🎙️✨ najva.brdkhsrv.com

Najva App @NajvaApp

5 months ago

1 2 5 5K 2

Download Image

3 12 65 5K 8

Download Image

CardioTechx @CardioTechx

6 months ago

🚨 New CardioTechX Journal Club Episode 🚨 Can an AI-enabled ECG detect diastolic dysfunction as well as echo? 📉 A novel deep learning model identifies LV filling pressures & diastolic grades with impressive accuracy (AUC up to 0.94) — even in echo-indeterminate cases.…

0 3 7 645 0

Download Video

CardioTechx @CardioTechx

6 months ago

🚨 CardioTechX Journal Club Episode🚨 In this episode we're highlighting a novel #MachineLearning model that identifies #AF patients that benefit most from #LAAO vs. #DOAC. Full manuscript found here: jacc.org/doi/full/10.10… #CardioTwitter #MedEd #Cardiology…