Pranav Dorbala

MD-PhD Student for genetics in personalized medicine @illinoisMED Carle Illinois College of Med @uofigrainger ECE PhD Student @georgiatech Biomedical Engineer

[David Ouyang, MD @David_Ouyang]

1/n We are excited to announce EchoPrime – the first echocardiography AI model capable of evaluating a full transthoracic echocardiogram study, identify the most relevant videos, and produce a comprehensive interpretation!

Great work lead by @milos_ai, EchoPrime is the largest and most complex #echofirst AI model yet, trained on 10x the data of existing models.

[Rohan Khera @rohan_khera]

In the 1st @JACCJournals issue w @hmkyale at the helm, proud to share our SOTA review on AI's transformative role in CV discovery & care 🧵👇

Great collaboration w colleagues & leaders.

@EricTopol @ekoikonomou @atulbutte @girish_nadkarni @jennawiens @jessRmorley

https://t.co/ynwWfNO1ZG

[David Ouyang, MD @David_Ouyang]

📢Excited to announce EchoCLIP, or vision-language foundation #AI model for #echofirst interpretation.

Using over 1,000,000 cardiac ultrasound videos paired with expert cardiologist interpretations, we enable #echofirst semantic search and zero-shot predictions.

EchoCLIP can zero-shot assess LVEF (MAE 7.1% on external validation), identify ICD, TAVR, MitraClip, and more!

1/🧵

[Jon Cunningham @JonWCunningham]

Thanks @VibhuParcha @jenniejlin @Circ_Gen for this video abstract

Breaking down our dicrotic notch paper on a whiteboard

@steven_lubitz @gpbatra @paolodkl

[Sam Rawal, MD @samarthrawal]

Spent some time exploring GPT-3 zero-shot capabilities of parsing patient medical records, including entity extraction, text summarization, and Question-Answering. Short thread of some highlights! #MedTwitter

Full notebook available at: https://t.co/UNZ0qjrj8g

[Partho P Sengupta @ppsengupta1]

Thank you, @adnanalkhouli. Congrats, @QuincyHathaway @YanamalaNaveena et al., for setting up "such a high standard..." as written in the editorial https://t.co/B6uERy7gvl Reverse translation proved "Ultrasomics" tracks myocyte geometry without needing manual measurements

[Jon Cunningham @JonWCunningham]

Join us today and tomorrow for the Biomarkers of the Future Symposium!

Excited to be speaking on proteomics in heart failure.

Registration is free
https://t.co/uyosDyFi3e

@AlexMebazaa @JJheart_doc @DrJasper1 @mvaduganathan

[National Human Genome Research Institute @genome_gov]

Researchers are increasingly using genomic data for diagnosing rare medical conditions, but current technology is not as accurate in predicting genomic risk for more common conditions such as heart disease and diabetes.

[David Ouyang, MD @David_Ouyang]

🚨🚨🚨🚨🚨🚨

We are REALLY excited to present the results of EchoNet-RCT, the first BLINDED RANDOMIZED clinical trial of #AI technology in cardiology at #ESCCongress 2022 in TWO days.

Late Breaking Clinical Trial Hot Line Session 3

Highlights of trial design include:

1/🧵

[Pranav Rajpurkar @pranavrajpurkar]

Now out 📢 Our work on deep learning to predict heart muscle injury with bedside monitors w/ Tom Jin, Dr. David Kim + more.

💡 We can use continuously recorded ECG to rule-out myocardial injury for a patient early in their emergency department visit

https://t.co/Hl4UionScL

[Pranav Rajpurkar @pranavrajpurkar]

Delighted to publicly share our 17+ lecture-long *AI Research Intensive*, designed to teach fundamental ML engineering skills and skills to write an AI research paper:

https://t.co/lRYpeZ7T5X

[Eric Green @NHGRI_Director]

One of the most common misperceptions about genomics and genetics is that genes determine everything in biology. But it is actually not genes alone! Phenotypes, or traits, are a result of complex interactions between genes and other biological factors, like the environment.

[European Society of Cardiology Journals @ESC_Journals]

Current state of #ArtificialIntelligence based algorithms for hospital admission prediction in patients with #HeartFailure - a scoping review https://t.co/A7NRRJqFL7

#EHJDigital #MachineLearning @BruiningNico @rafavidalperez @pauldub67

[Diego Ramonfaur @d_ramonfaur]

Honored to have been the student speaker for @HarvardMed's graduation ceremony. I'm proud of all my peers in the class of 2022 and looking forward to hearing more about your future success.

[Avraham Z. Cooper, MD 🩺 @AvrahamCooperMD]

What constrictive pericarditis ACTUALLY LOOKS LIKE

Via @omorenou @grepmeded

#MedTwitter

[MIT CSAIL @MIT_CSAIL]

Four deep learning papers to read right now: https://t.co/YSCof14ija

credit @RobertTLange

[Diego Ramonfaur @d_ramonfaur]

Successfully defended my dissertation at @harvardmed.

Incredibly grateful with my family, mentor, and peers that were part of this learning experience.

[Avraham Z. Cooper, MD 🩺 @AvrahamCooperMD]

This is still the best example of sustained clonus I’ve ever seen (serotonin syndrome)

Via @DavidJuurlink @grepmeded

#MedTwitter

[Monica Agrawal @MonicaNAgrawal]

How can we pre-train good representations of longitudinal health data?

Scramble a patient’s data (e.g radiology notes), and ask your model to unscramble! For chronic conditions, the model learns to attend to features corresponding to disease progression