PhysioNet

Featured Resources

Software Open Access

Software for computing Heart Rate Fragmentation

Madalena Costa

Heart rate fragmentation: a new method for the analysis of cardiac interbeat interval time series. The code provided can be run in Windows, Mac and Linux machines.

heart rate variability aging cardiovascular disease vagal tone time series analysis prediction of atrial fibrillation cardiac autonomic function prediction of cardiovascular events prediction of cognitive decline heart rate fragmentation

Published: Feb. 14, 2024. Version: 1.0.0

Challenge Credentialed Access

SNOMED CT Entity Linking Challenge

Will Hardman, Mark Banks, Rory Davidson, Donna Truran, Nindya Widita Ayuningtyas, Hoa Ngo, Alistair Johnson, Tom Pollard

272 discharge notes from the MIMIC-IV-Note dataset annotated with SNOMED CT concepts.

snomed entity linking clinical annotation

Published: Dec. 19, 2023. Version: 1.0.0

Database Open Access

VitalDB, a high-fidelity multi-parameter vital signs database in surgical patients

Hyung-Chul Lee, Chul-Woo Jung

VitalDB, a high-fidelity multi-parameter vital signs database in surgical patients

waveform anesthesia vitaldb intraoperative biosignal ecg

Published: Sept. 21, 2022. Version: 1.0.0

Database Credentialed Access

MIMIC-IV

Alistair Johnson, Lucas Bulgarelli, Tom Pollard, Steven Horng, Leo Anthony Celi, Roger Mark

Large database of de-identified health information from patients admitted to Beth Israel Deaconess Medical Center

mimic critical care machine learning intensive care unit

Published: Jan. 6, 2023. Version: 2.2

Database Credentialed Access

MIMIC-CXR Database

Alistair Johnson, Tom Pollard, Roger Mark, Seth Berkowitz, Steven Horng

Chest radiographs in DICOM format with associated free-text reports.

mimic computer vision machine learning chest x-rays radiology natural language processing

Published: Sept. 19, 2019. Version: 2.0.0

Database Credentialed Access

BRAX, a Brazilian labeled chest X-ray dataset

Eduardo Pontes Reis, Joselisa Paiva, Maria Carolina Bueno da Silva, Guilherme Alberto Sousa Ribeiro, Victor Fornasiero Paiva, Lucas Bulgarelli, Henrique Lee, Paulo Victor dos Santos, vanessa brito, Lucas Amaral, Gabriel Beraldo, Jorge Nebhan Haidar Filho, Gustavo Teles, Gilberto Szarf, Tom Pollard, Alistair Johnson, Leo Anthony Celi, Edson Amaro

BRAX contains 24,959 chest radiography exams and 40,967 images acquired in a large general Brazilian hospital. All images have been read by trained radiologists and 14 labels were derived from Brazilian Portuguese reports using NLP.

chest x-ray dataset artificial intelligence

Published: June 17, 2022. Version: 1.1.0

Latest Resources

Database Open Access

Gesture Recognition and Biometrics ElectroMyogram (GRABMyo)

Ning Jiang, Ashirbad Pradhan, Jiayuan He

Open-access dataset of electromyogram (EMG) recordings collected from the wrist and forearm muscles of 43 people while they performed hand gestures.

Published: June 7, 2024. Version: 1.1.0

Visualize waveforms

Model Credentialed Access

Me-LLaMA: Foundation Large Language Models for Medical Applications

Qianqian Xie, Qingyu Chen, Aokun Chen, Cheng Peng, Yan Hu, Fongci Lin, Xueqing Peng, Jimin Huang, Jeffrey Zhang, Vipina Keloth, Xinyu Zhou, Huan He, Lucila Ohno-Machado, Yonghui Wu, Hua Xu, Jiang Bian

Me-LLaMA is a family of large language models for medical applications trained using clinical text with LLaMA2 models as the base. We release model weights for the foundation models as well as the chat-enhanced models.

large language models

Published: June 5, 2024. Version: 1.0.0

Database Credentialed Access

A Temporal Dataset for Respiratory Support in Critically Ill Patients

Mira Moukheiber, Lama Moukheiber, Dana Moukheiber, Sicheng Hao, Leo Anthony Celi, Hyung-Chul Lee

A benchmark dataset offering hourly records over a 90-day period for 50,920 ICU subjects, including dynamic pulmonary function data and a spectrum of covariates for respiratory intervention analyses.

oberservational data time-series

Published: May 31, 2024. Version: 1.0.0

Database Credentialed Access

MIMIC-IV-Ext Clinical Decision Making: A MIMIC-IV Derived Dataset for Evaluation of Large Language Models on the Task of Clinical Decision Making for Abdominal Pathologies

Paul Hager, Friederike Jungmann, Daniel Rueckert

A curated set of ED clinical decision making cases for four abdominal pathologies. Each case contains the exams required to diagnose including HPI, physical examination, laboratory tests, and imaging. Relevant treatment information is also included.

clinical decision making emergency room diagnosis abdominal pathologies treatment plan large language models

Published: May 17, 2024. Version: 1.0

Database Restricted Access

DREAMT: Dataset for Real-time sleep stage EstimAtion using Multisensor wearable Technology

Ke Wang, Jiamu Yang, Ayush Shetty, Jessilyn Dunn

Dataset for Real-time sleep stage EstimAtion using Multisensor wearable Technology

wearable biomedical sleep disorders time series classification

Published: April 30, 2024. Version: 1.0.0

Database Credentialed Access

Medical Expert Annotations of Unsupported Facts in Doctor-Written and LLM-Generated Patient Summaries

Stefan Hegselmann, Shannon Shen, Florian Gierse, Monica Agrawal, David Sontag, Xiaoyi Jiang

Annotations for unsupported facts in 100 original MIMIC patient summaries (discharge instructions) and hallucinations in 100 Large Language Model (LLM) generated patient summaries labeled by two medical experts.

Published: April 28, 2024. Version: 1.0.0

More resources

News

George B. Moody PhysioNet Challenge 2024 is teaming up with Data Science Africa

May 18, 2024

The George B. Moody PhysioNet Challenge 2024 is teaming up with Data Science Africa (DSA) to innovate cardiac care through signal processing & machine learning at Nyeri, Kenya, for DSA's 10 year anniversary in June 2024.

DSA aims to create a hub in the network of data science researchers across Africa and strengthen the African data science community. #DSA2024Nyeri https://www.datascienceafrica.org/dsa2024nyeri/blog/summer-school

Guidelines for creating datasets and models from MIMIC

News from: MIMIC-IV v2.2.

April 24, 2024

We recognize that there is value in creating datasets or models that are either derived from MIMIC or which augment MIMIC in some way (for example, by adding annotations). Here are some guidelines on creating these datasets and models:

Any derived datasets or models should be treated as containing sensitive information. If you wish to share these resources, they should be shared on PhysioNet under the same agreement as the source data.
If you would like to use the MIMIC acronym in your project name, please include the letters “Ext” (for example, MIMIC-IV-Ext-YOUR-DATASET"). Ext may either indicate “extracted” (e.g. a derived subset) or “extended” (e.g. annotations), depending on your use case.

Network issues at MIT, impacting the availability of PhysioNet

April 9, 2024

We are currently experiencing network issues at MIT, impacting the availability of PhysioNet services. We apologize for any inconvenience this may cause and are working to resolve the issue as soon as we can.

George B. Moody PhysioNet Challenge 2024: Challenge Update

March 14, 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. In connection with this, the Challenge organizers will be running a workshop at DSA in Kenya from June 2-5th 2024. Please note that we are also accepting (and scoring) entries, and there are two deadlines coming up - April 8th to submit a preliminary entry to the Challenge and April 15th to submit a (placeholder) abstract to CinC.

Duke Critical Care Datathon: 13-14 April 2024

Feb. 7, 2024

Our colleagues at Duke are hosting a Critical Care Datathon on April 13-14, 2024. The Datathon is a collaborative two-day event that connects critical care clinicians with data scientists to develop pragmatic data-driven models using de-identified critical care electronic health record datasets.

Using de-identified critical care electronic health record datasets (including MIMIC and the eICU Collaborative Research Database), participants will develop new projects in 36 hours, from problem to abstract (and more)!

Participants will be organized into teams that are half-data science, half-clinical. You do not need to have a team; the organizers will help you find a team. Questions will be crowdsourced. No experience is required.

If you are a clinician, your interest, but not expertise, in data science is required.
If you are a data scientist, your interest, but not expertise, in healthcare and critical care is required.

For more information, see: https://sites.duke.edu/datathon2024/

More news