Resources

A curated collection of notable AI research papers on physiological signals.

A curated list of foundation models, datasets and tools for biosignals.

An open platform for evaluating seizure detection algorithms on EEG, built around the SzCORE evaluation framework.

A long-running repository of freely available physiological and clinical datasets, with the software to work with them.

A platform for sharing large clinical brain datasets, such as EEG and sleep recordings, for research.

A large collection of clinical EEG recordings from Temple University Hospital, released with annotated subsets.

A free platform for sharing neuroimaging and electrophysiology datasets in a standard format.

A Python package for exploring, visualising and analysing EEG and MEG data.

A Python toolbox for processing physiological signals such as ECG, PPG, EDA, EMG and respiration.

A Python library for reading, writing and processing the waveform formats used by PhysioNet.

Updated guidance for reporting clinical prediction models that use regression or machine learning.

Reporting guidance for clinical trials of interventions that involve artificial intelligence.

Guidance for writing protocols of clinical trials of interventions that involve artificial intelligence.

Transparent Reporting of Ethics for Generative AI: a checklist for assessing and reporting the ethical aspects of generative AI research in health care.

A short commentary on how benchmark results in biomedical machine learning can mislead.

Principles for moving clinical AI from benchmark results to real-world evaluation.

A commentary asking what evidence shows that clinical AI improves care.

A comparison of ECG foundation models across clinical tasks, and where benchmark claims fall short.

A survey of foundation models for biosignals, their training objectives and downstream tasks.

An account of PhysioNet, the open platform behind many public biosignal datasets.