Launching the official phase for the PhysioNet Challenge 2023

[PhysioNet Challenge]

Dear Challengers,

We are pleased to announce the beginning of the official phase of the George B. Moody PhysioNet Challenge 2023: Predicting Neurological Recovery from Coma After Cardiac Arrest.

For the official phase of this year’s Challenge, we have dramatically increased the size of the Challenge dataset, which now includes

• EEG and ECG data, including many additional channels;
• full recordings instead of 5-minute windows from each hour of a recording;
• a reference montage representation for the EEG channels instead of a sequential montage representation, which allows you to decide how to represent the EEG data for your models; and
• (de-identified) hospital identifiers.

We are still uploading the training set to PhysioNet, but we expect to post it in the coming days.

For better or worse, these additions have increased the size of the data by over an order of magnitude. Please check the Challenge website for how to use wget to download the data, including various subsets of the training set:
https://physionetchallenges.org/2023/#data

We have updated the example code to use the new data:
https://github.com/physionetchallenges/python-example-2023
https://github.com/physionetchallenges/matlab-example-2023

We have also updated the evaluation code, which now computes the true positive rate of your model at a false positive rate of 0.05 for each hospital instead of a false positive rate of 0.05 across the entire (split of the) dataset:
https://github.com/physionetchallenges/evaluation-2023

The Challenge task – predict whether coma patients have good or poor recoveries following cardiac arrest – remains the same. Please see the Challenge webpage for details, including links to the new data and a new submission form for when you’re ready:
https://physionetchallenges.org/2023/

As always, there may be minor changes to the data and code as we launch and relaunch, so please stay tuned for updates.

The reviews of your Computing in Cardiology abstracts are nearly finished. We look forward to sharing the results with you in the coming weeks, and we look forward to welcoming you to Atlanta in a few short months.

Best,

The PhysioNet Challenge Organizers

Please post questions and comments in the forum. However, if your question reveals information about your entry, then please email info at physionetchallenge.org. We may post parts of our reply publicly if we feel that all Challengers should benefit from it. We will not answer emails about the Challenge to any other address. This email is maintained by a group. Please do not email us individually.

[PhysioNet Challenge]

Dear Challengers,

We have now posted the updated training set for the official phase of the 2023 Challenge:

https://physionetchallenges.org/2023/#data

https://physionet.org/content/i-care/

Due to the large size (1.6 TB) of the training set, we may make a few minor changes in the coming days, but please feel free to start using it to develop your models. If you only want to download a subset of the data (e.g., the first 72 hours of each patient's data in the training set are *only* 315 GB), then please see the example commands in this section of the Challenge webpage:

https://physionetchallenges.org/2023/#accessing-data

Best,

Matt

Please post questions and comments in the forum. However, if your question reveals information about your entry, then please email info at physionetchallenge.org. We may post parts of our reply publicly if we feel that all Challengers should benefit from it. We will not answer emails about the Challenge to any other address. This email is maintained by a group. Please do not email us individually.