📢 [Models]
Deep learning models have shown remarkable performance in electrocardiogram (ECG) analysis, but their success has been constrained by the limited availability and size of ECG datasets, resulting in systems that are more task specialists than versatile generalists. In this work, we introduce HuBERT-ECG, a foundation ECG model pre-trained in a self-supervised manner on a large and diverse dataset of 9.1 million 12-lead ECGs encompassing 164 cardiovascular conditions. By simply adding an output layer, HuBERT-ECG can be fine-tuned for a wide array of downstream tasks, from diagnosing diseases to predicting future cardiovascular events. Across diverse real-world scenarios, HuBERT-ECG achieves AUROCs from 84.3% in low-data settings to 99% in large-scale setups. When trained to detect 164 overlapping conditions simultaneously, our model delivers AUROCs above 90% and 95% for 140 and 94 diseases, respectively. HuBERT-ECG also predicts death events within a 2-year follow-up with an AUROC of 93.4%. We release models and code.
- [12/2024] Reproducibility has never been easier! Training, validation, and test splits ready to use in the reproducibility folder!
- [12/2024] Pre-trained models are easily downloadable from Hugging Face using
AutoModel.from_pretrained - [11/2024] Pre-trained models and models fine-tuned on Cardio-Learning are freely available on HuggingFace
- [11/2024] This repository has been made public!
All our models are accessible on Hugging Face [(https://huggingface.co/Edoardo-BS/HuBERT-ECG/)] under CC BY-NC 4.0 license
Clone this repository and install all the necessary dependecies written in the requirements.txt file with pip install -r requirements.txt.
Full installation time may take up to 1 minute.
In the reproducibility folder you can find all train, validation, and test splits we used in our work as .csv files. You simply have to follow the instructions in the reproducibility/README.md to reproduce our results.
As an example, you can easily fine-tune and evaluate an instance of HuBERT-ECG on PTB-XL All dataset, as shown in .sh scripts finetune.sh and test.sh.
Prediction on a single instance takes less than 1 second on an A100 GPU node.
If you use our models or find our work useful, please consider citing us:
doi: https://doi.org/10.1101/2024.11.14.24317328