George B. Moody PhysioNet Challenges

For the past 25 years, PhysioNet and Computing in Cardiology have co-hosted a series of annual challenges, now called the George B. Moody PhysioNet Challenges, to tackle clinically interesting but unsolved questions.

The George B. Moody PhysioNet Challenge 2024 invites teams to develop algorithms for digitizing and classifying electrocardiograms (ECGs) captured from images or paper printouts. Despite recent advances in digital ECG devices, paper or physical ECGs remain common, especially in the Global South. These paper ECGs document the history and diversity of cardiovascular diseases (CVDs), and algorithms that can digitize and classify these images have the potential to improve our understanding and treatment of CVDs, especially for underrepresented and underserved populations.

We ask participants to design and implement working, open-source algorithms that, based only on the provided ECG images, reconstruct the waveforms and/or classify or diagnose the images. The teams with the best scores for these tasks on the hidden test set win the Challenge.

Please check the below links for information about current and past Challenges, including important details about scoring and test data for previous Challenges.

Select News

September 20, 2024: We have released (and updated) the results of the 2024 Challenge. Congratulations to the winners! Please see the announcement for more details.
August 19, 2024: We have extended the official phase deadline to 23:59 GMT on August 21, 2024. Please see the deadlines for the updated deadlines.
July 9, 2024: The PhysioNet Challenges spent several days last month at Data Science Africa 2024. The DSAIL at DeKUT helped us to run a hackathon over 4 days with more than 50 attendees, including some that we will win a prize from the IEEE SPS at CinC 2024. Please see this announcement for more information about our trip to DSA.
June 14, 2024: Abstract decisions for CinC 2024 have been announced, and the official scoring system is open with a new validation data set that includes images of real paper ECGs. Please submit entries early and often; although you have 10 entries in the scoring system, we may limit each team to one request or deprioritize additional entries from a team in the final week because of the large numbers of entries that we receive in the final days of the Challenge, and because teams that enter earlier tend to perform better!
June 6, 2024: We have just completed a 3-day workshop at this year’s annual DSA meeting in Nyeri, Kenya. As part of this capacity-building initiative, we provided another working solution to the ECG image classification task. The code for this can be found here. The Challenge teams are welcome to use any part of this code in their entry, but we strongly suggest trying other approaches as well. In particular, we deliberately under-designed the model, which does not attempt to digitize the data for classification. We have posted this only to provide a working example, and not as a high bar to beat.
May 24, 2024: The official phase of the George B. Moody PhysioNet Challenge 2024 has begun! We have made many updates to make the Challenge more accessible and more realistic. Please see our announcement on the Challenge forum for more details and begin submitting your updated entries in the coming days.
March 15, 2024: We are delighted to announce that the George B. Moody PhysioNet Challenges are partnering with Data Science Africa (DSA) and the IEEE Signal Processing Society’s Challenges and Data Collections Committee (CDCC). The IEEE CDCC is supporting this year’s Challenge with additional cash prizes for participating teams from Africa, and the Challenge organizers will be running a workshop at this year’s annual DSA meeting in Kenya from June 2-5, 2024. Please note that we are also accepting (and scoring) entries, and there are two deadlines coming up: April 8, 2024 to submit a preliminary entry to the Challenge and April 15, 2024 to submit a (placeholder) abstract to CinC.
February 29, 2024: We are now accepting unofficial phase submissions for the 2024 Challenge. Please read the submissions instructions, double check your code, and submit your code when ready.
January 25, 2024: The NIH-funded George B. Moody PhysioNet Challenge 2024 is now open! Please read this website for details and share questions and comments on Challenge forum. This year’s Challenge is generously sponsored by MathWorks and AWS.
October 10, 2023: The winners of the 2023 Challenge were announced on 4 October 2023 at CinC 2023 in Atlanta, GA, USA. Congratulations, teams! See the results and papers as well as the full announcement for the final steps in this year’s Challenge.
… see previous news articles here.

Current Challenge

2024: Digitization and Classification of ECG Images: The George B. Moody PhysioNet Challenge 2024

Past Challenges

2023: Predicting Neurological Recovery from Coma After Cardiac Arrest
Paper and contributed software collection ongoing
2022: Heart Murmur Detection from Phonocardiogram Recordings
44+ papers and 77 contributed software
2021: Will Two Do? Varying Dimensions in Electrocardiography
60+ papers and 58 contributed software
2020: Classification of 12-lead ECGs
62+ papers and 70 contributed software
2019: Early Prediction of Sepsis from Clinical Data
55+ papers and 91 contributed software
2018: You Snooze, You Win
23 papers and 19 contributed software
2017: AF Classification from a Short Single Lead ECG Recording
57 papers and 64 contributed software
2016: Classification of Normal/Abnormal Heart Sound Recordings
11 papers and 48 contributed software
2015: Reducing False Arrhythmia Alarms in the ICU
20 papers and 28 contributed software
2014: Robust Detection of Heart Beats in Multimodal Data
15 papers and 35 contributed software
2013: Noninvasive Fetal ECG
29 papers and 17 contributed software
2012: Predicting Mortality of ICU Patients
17 papers and 58 contributed software
2011: Improving the Quality of ECGs Collected using Mobile Phones
17 papers and 7 contributed software
2010: Mind the Gap
13 papers and 5 contributed software
2009: Predicting Acute Hypotensive Episodes
11 papers and 4 contributed software
2008: Detecting and Quantifying T-Wave Alternans
19 papers and 5 + 1 contributed software
2007: Electrocardiographic Imaging of Myocardial Infarction
6 papers
2006: QT Interval Measurement
20 papers and 6 contributed software
2005: The First Five Challenges Revisited
5 papers
2004: Spontaneous Termination of Atrial Fibrillation
12 papers and 1 contributed software
2003: Distinguishing Ischemic from Non-Ischemic ST Changes 3 papers and 1 contributed software
2002: RR Interval Time Series Modeling
12 papers and 10 contributed software
2001: Predicting Paroxysmal Atrial Fibrillation
9 papers
2000: Detecting Sleep Apnea from the ECG
13 papers and 1 contributed software

Supported by the National Institute of Biomedical Imaging and Bioengineering (NIBIB) under NIH grant number R01EB030362.