Spaces:
Sleeping
Sleeping
add doi
Browse files
README.md
CHANGED
|
@@ -10,4 +10,49 @@ pinned: false
|
|
| 10 |
license: apache-2.0
|
| 11 |
---
|
| 12 |
|
| 13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
license: apache-2.0
|
| 11 |
---
|
| 12 |
|
| 13 |
+
# ECG Classification with Self-Supervised Learning
|
| 14 |
+
|
| 15 |
+
[](https://doi.org/10.57967/hf/8469)
|
| 16 |
+
|
| 17 |
+
**Interactive Space** for ECG cardiovascular disease classification using SimCLR pre-trained model fine-tuned on PTB-XL.
|
| 18 |
+
|
| 19 |
+
## Quick Start
|
| 20 |
+
|
| 21 |
+
Upload an ECG file in any format (DICOM, MATLAB, HDF5, WFDB, CSV, and more) and get instant predictions across 5 diagnostic classes.
|
| 22 |
+
|
| 23 |
+
## Model Performance
|
| 24 |
+
|
| 25 |
+
- **AUROC:** 0.8717
|
| 26 |
+
- **Accuracy:** 0.8234
|
| 27 |
+
- **Training Data:** PTB-XL with 10% labeled samples (1,747 ECGs)
|
| 28 |
+
|
| 29 |
+
## Supported File Formats
|
| 30 |
+
|
| 31 |
+
- **Clinical:** DICOM (.dcm), SCP-ECG (.scp), HL7 aECG (.xml)
|
| 32 |
+
- **Research:** WFDB (.hea/.dat), EDF (.edf)
|
| 33 |
+
- **Scientific:** MATLAB (.mat), HDF5 (.h5), NumPy (.npy)
|
| 34 |
+
- **Generic:** CSV/TXT/TSV, Binary (.raw, .bin)
|
| 35 |
+
|
| 36 |
+
## Citation
|
| 37 |
+
|
| 38 |
+
If you use this model, please cite it using the DOI:
|
| 39 |
+
|
| 40 |
+
```
|
| 41 |
+
10.57967/hf/8469
|
| 42 |
+
```
|
| 43 |
+
|
| 44 |
+
**Bibtex:**
|
| 45 |
+
```bibtex
|
| 46 |
+
@model{ssrl_ecg_2024,
|
| 47 |
+
title={SSL ECG Classification},
|
| 48 |
+
doi={10.57967/hf/8469},
|
| 49 |
+
url={https://huggingface.co/spaces/Tumo505/SSL-ECG-Classification},
|
| 50 |
+
year={2024}
|
| 51 |
+
}
|
| 52 |
+
```
|
| 53 |
+
|
| 54 |
+
## Resources
|
| 55 |
+
|
| 56 |
+
- [Model Card](https://huggingface.co/Tumo505/SSL-ECG-Classification-model-card)
|
| 57 |
+
- [GitHub Repository](https://github.com/Tumo505/SSL-for-ECG-classification)
|
| 58 |
+
- [Model Weights](https://huggingface.co/Tumo505/SSL-ECG-Classificcation)
|
app.py
CHANGED
|
@@ -601,6 +601,8 @@ with gr.Blocks(
|
|
| 601 |
|
| 602 |
### About This Model
|
| 603 |
|
|
|
|
|
|
|
| 604 |
**Architecture:** 1D CNN with SimCLR self-supervised pre-training
|
| 605 |
|
| 606 |
**Training:**
|
|
@@ -615,9 +617,6 @@ with gr.Blocks(
|
|
| 615 |
- CD: Conduction Disturbances
|
| 616 |
|
| 617 |
**Research Only** - Not validated for clinical use
|
| 618 |
-
|
| 619 |
-
[View Model Card](https://huggingface.co/Tumo505/SSL-ECG-Classification-model-card)
|
| 620 |
-
[GitHub Repository](https://github.com/Tumo505/SSL-for-ECG-classification)
|
| 621 |
""")
|
| 622 |
|
| 623 |
|
|
|
|
| 601 |
|
| 602 |
### About This Model
|
| 603 |
|
| 604 |
+
**DOI:** [`10.57967/hf/8469`](https://doi.org/10.57967/hf/8469) | [Model Card](https://huggingface.co/Tumo505/SSL-ECG-Classification-model-card) | [GitHub](https://github.com/Tumo505/SSL-for-ECG-classification)
|
| 605 |
+
|
| 606 |
**Architecture:** 1D CNN with SimCLR self-supervised pre-training
|
| 607 |
|
| 608 |
**Training:**
|
|
|
|
| 617 |
- CD: Conduction Disturbances
|
| 618 |
|
| 619 |
**Research Only** - Not validated for clinical use
|
|
|
|
|
|
|
|
|
|
| 620 |
""")
|
| 621 |
|
| 622 |
|