Видео

Can you share the code

Nice content you're really doing great

This is amazing!

You should spend more time on explaining the model itself because that's what is the most interesting. You just showed a picture without any explanation.

could you share the slides please?

This could be the dumbest question you get, but how does someone take a trained model after using PyHealth and turn it into something like an API to call and get predictions from?

thanks!

Hi Thank you for sharing this! I wonder if it’s worked for MIMIC-IV. Thank you!

'PromoSM' 💥

Nice work, where can I get access to these slides

This is awesome! Thank you guys

Content: 00:00 - Intro 00:54 - PyHealth Overview 03:25 - Live Session Schedule 04:41 - 5-stage Pipeline 05:58 - Useful Resources

Outline: 00:00 - Intro 01:09 - Customized SampleDataset Example 05:14 - Summary of the Live 07:21 - Useful resources

Outline: 00:00 - Intro (slides) 00:22 - Intro (Colab) 01:24 - Usecase example 02:20 - Task example

Outline: 00:00 - Overview of this Live (slides) 02:00 - Instructions for pyhealth.datasets (Colab) 04:46 - Examples to process MIMIC-III 06:27 - Query information after processing 09:11 - Examples to process OMOP-CDM datasets

Outline: 00:00 - Intro 01:40 - Data structure (slides) 04:18 - Event structure 05:40 - Visit structure 08:08 - Patient structure

Outline: ------------------------------ 00:00 - Introduction 03:58 - Overview of PyHealth 09:12 - Key designs in PyHealth 13:24 - PyHealth installation 13:48 - 5-stage ML pipeline 15:20 - Example 1: Relieve the pain of data processing 20:04 - Example 2: Build the model for you 23:13 - Example 3: Master the medical coding systems 26:57 - Summary of the PyHealth 28:04 - Development Team and Acknowledgements 28:36 - PyHealth resources and call for contributions

Terrific interpretation. JT-VAE has been so popular that you would like to come across this approach in every summary paper regarding molecule generation. For the first time I almost get the essence behind it.

Can you explain how can you find dimension-wise probability?

excellent and important work; typical of your and your team's work and contributions. We are trying to adopt your work to predict adverse events in cancer patients. We downloaded your code and it works fine using the mimic3 dataset as input. We have successfully run the trained model and the test one with no errors.

Man, such awesome video, sooo under-rated

Please share with me your email i need your advise

Thank you brother

what is there in the embedding module and transforming module, and where did you get the data for interaction module ?

🤤 Promo-SM!!

Hello Professor, Is this book required for the CSE6250 for OMSA in Georgia Tech? I am planning to take the course in Spring2022 semester and was thinking to get prepared for the course.

Dear professor, Great piece of work. Curious to know how handy it is in terms of looking into relevant jupyter notebooks or repository for hands on case studies. Thank you.

It's a very interesting work with high potentials. However, its implementation code is poorly written with many bugs. We have managed to resolve some of the bugs but not all. Unfortunately, the first author is not responding to our emails. It'd be great if you can help us reach the first author, we only wish to test in our hospital setting. Thanks.

Paper: arxiv.org/abs/1911.10395

Thanks for the explaining

Did anyone implemented this paper ? If so please send the google Colab link. Thank U.

Hello Jimeng, This is Soumava Dey, a student of your DLH course, just subscribed to your channel. Videos look very good and informative. I do have my own RUclips channel on Data Science and I will definitely share your RUclips channel link on my platform for better reach.

nice quick review of RNN (recurrent neural networks) @ 1:18 & GRU (gated recurrent unit) @ 5:55 Doctor AI paper coverage proper starts @ 11:12

Milky makes a cameo at the beginning

Thank you professor Jim Sun. I did you Big Data For Health and one of best course in GT. Thank you for sharing.

Related papers: sleep.csail.mit.edu/files/rfsleep-paper.pdf www.ijcai.org/Proceedings/2019/0817.pdf

This is the related video about STAN paper ruclips.net/video/t6BAAzk6ZK0/видео.html&ab_channel=AIforhealth

Here is the paper. academic.oup.com/jamia/advance-article/doi/10.1093/jamia/ocaa322/6118380

SkipGNN code github.com/kexinhuang12345/SkipGNN

SkipGNN paper www.nature.com/articles/s41598-020-77766-9

MolTrans code link github.com/kexinhuang12345/MolTrans

MolTrans paper link academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/btaa880/5929692

Thanks for this!

hi