Skip to main content

deepfake-ecg

Paper | GitHub | Pre-generated ECGs (150k)

Generate unlimited realistic deepfake ECGs using the deep generative model:Pulse2pulse introduced in our full paper here: https://doi.org/10.1101/2021.04.27.21256189 (DeepFake electrocardiograms: the key for open science for artificial intelligence in medicine)

Installation

Use the package manager pip to install deepfake-ecg generator.

pip install deepfake-ecg

Usage

The generator functions can generate DeepFake ECGs with 8-lead values [lead names from first coloum to eighth colum: 'I','II','V1','V2','V3','V4','V5','V6'] for 10s (5000 values per lead). These 8-leads format can be converted to 12-leads format using the following equations.

lead III value = (lead II value) - (lead I value)
lead aVR value = -0.5*(lead I value + lead II value)
lead aVL value = lead I value - 0.5 * lead II value
lead aVF value = lead II value - 0.5 * lead I value

Run on CPU (default setting)

import deepfakeecg

#deepfakeecg.generate("number of ECG to generate", "Path to generate", "start file ids from this number", "device to run") 

deepfakeecg.generate(5, ".", start_id=0, run_device="cpu") # Generate 5 ECGs to the current folder starting from id=0

Run on GPU

import deepfakeecg

#deepfakeecg.generate("number of ECG to generate", "Path to generate", "start file ids from this number", "device to run") 

deepfakeecg.generate(5, ".", start_id=0, run_device="cuda") # Generate 5 ECGs to the current folder starting from id=0

Pre-generated DeepFake ECGs and corresponding MUSE reports are here: https://osf.io/6hved/

- In this repository, there are two DeepFake datasets:
    1. 150k dataset - Randomly generated 150k DeepFakeECGs
    2. Filtered all normals dataset - Only "Normal" ECGs filtered using the MUSE analysis report

A real ECG vs a DeepFake ECG (from left to right):

GitHub Logo

A sample DeepFake ECG:

GitHub Logo

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

Please make sure to update tests as appropriate.

Citation:

@article{ecg-pulse2pulse,
	author = {Thambawita, Vajira Lasantha and Isaksen, Jonas L and Hicks, Steven and Ghouse, Jonas and Ahlberg, Gustav and Linneberg, Allan and Grarup, Niels and Ellervik, Christina and Olesen, Morten Salling and Hansen, Torben and Graff, Claus and Holstein-Rathlou, Niels-Henrik and Str{\"u}mke, Inga and Hammer, Hugo L. and Maleckar, Mary M and Halvorsen, P{\aa}l and Riegler, Michael A. and Kanters, J{\o}rgen K.},
	doi = {10.1101/2021.04.27.21256189},
	elocation-id = {2021.04.27.21256189},
	journal = {medRxiv},
	publisher = {Cold Spring Harbor Laboratory Press},
	title = {DeepFake electrocardiograms: the key for open science for artificial intelligence in medicine},
	url = {https://doi.org/10.1101/2021.04.27.21256189},
	year = {2021}
}

License

MIT

For more details:

Please contact: vajira@simula.no, michael@simula.no

Release files for deepfake-ecg 1.1.2

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for deepfake-ecg 1.1.2
File Size Uploaded
deepfake-ecg-1.1.2.tar.gz 39.4 MB Details

Built distribution (wheel)

Table of built distributions (wheels) for deepfake-ecg 1.1.2
File Interpreter ABI Platform
deepfake_ecg-1.1.2-py3-none-any.whl Python 3 none any Details

Total release size: 78.8 MB

Release files / deepfake-ecg-1.1.2.tar.gz

Download URL deepfake-ecg-1.1.2.tar.gz
Size 39.4 MB
Tags Source
SHA-256 checksum
How to use checksums
490f63705acb167c3ecf02f5a83e9147b844b9d1643e2b4ba00fa1c48281cfa0
BLAKE2b-256 checksum
How to use checksums
5737b923e8d863a30e859a1d4bacb93a4f5328c025dd6494b3ad68854befd647
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/3.3.0 pkginfo/1.6.1 requests/2.25.1 setuptools/51.1.2 requests-toolbelt/0.9.1 tqdm/4.55.2 CPython/3.8.2

Release files / deepfake_ecg-1.1.2-py3-none-any.whl

Download URL deepfake_ecg-1.1.2-py3-none-any.whl
Size 39.4 MB
Tags Python 3
SHA-256 checksum
How to use checksums
dff913de571f9d92d95e25eed4a3fe7df0bfee7c66315ba1fa0687acdd11649c
BLAKE2b-256 checksum
How to use checksums
36d406f4457cafcb4d3e57c5c98bc1a2a82ff0feac66286cd63ca3ad88c1d446
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/3.3.0 pkginfo/1.6.1 requests/2.25.1 setuptools/51.1.2 requests-toolbelt/0.9.1 tqdm/4.55.2 CPython/3.8.2

Release history Release notifications | RSS feed

This release

1.1.2 This release

2 release files

1.1.1

2 release files

1.1.0

2 release files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page