YerevaNN / mimic3-benchmarks

Python suite to construct benchmark machine learning datasets from the MIMIC-III 💊 clinical database.
https://arxiv.org/abs/1703.07771
MIT License
796 stars 328 forks source link

Waveforms #59

Open Saqibm128 opened 6 years ago

Saqibm128 commented 6 years ago

Are there any plans to try to integrate the high frequency waveform data from the MIMIC3 Waveform Dataset into the benchmark somehow? Would there be any interest?

Would this be likely out of scope of the project for now?

https://physionet.org/physiobank/database/mimic3wdb/

turambar commented 6 years ago

Hey @Saqibm128 we have no immediate plans to do this ourselves: expanding coverage of variables, incorporating medications and treatments, and adding additional prediction tasks, e.g., sepsis, are higher priorities. Also, incorporating clinical notes.

We would happily accept a pull request concerning waveforms -- my main advice is to couple this to a specific benchmark prediction task so that non-expert machine learners know when and how to use the waveforms appropriately. The vision of this project is to define specific tasks and data sets so clearly that machine learners can work with them without extensive previous experience with clinical data. That's how we add value on top of MIMIC.

tahakucukkatirci commented 4 years ago

adding additional prediction tasks, e.g., sepsis, are higher priorities.

Hi, is there any update related to sepsis prediction? Do you have a plan to add it in near future? @turambar @hrayrhar

Hrant-Khachatrian commented 4 years ago

No plans in the near future. Thanks for your interest