Improve model card: Add paper link, pipeline tag, specific license, and sample usage

This PR enhances the model card for the EchoingECG model by:

* **Updating Metadata**:
* Adding `pipeline_tag: other` for better discoverability of this specialized model.
* Correcting the `license` to `cc-by-nc-nd-4.0` to match the explicit license statement in the GitHub repository.
* Adding `medical` to the list of `tags` for improved categorization.
* **Improving Content**:
* Adding a direct link to the Hugging Face paper page: [EchoingECG: An Electrocardiogram Cross-Modal Model for Echocardiogram Tasks](https://huggingface.co/papers/2509.25791).
* Reformatting the GitHub repository link for better readability.
* Including the "Features" section from the GitHub README for a more comprehensive overview.
* Updating the image link to an absolute URL for robustness.
* Adding a detailed "Quick Start: Run EchoingECG in Jupyter Notebook" section with the code snippet from the GitHub README to provide clear sample usage.
* Including the full "License" details directly from the GitHub repository for transparency.

These changes aim to provide more accurate and complete information, improving the model's clarity and usability on the Hugging Face Hub.

README.md

@@ -1,19 +1,29 @@
 ---
-license: cc-by-4.0
 language:
 - en
+license: cc-by-nc-nd-4.0
 tags:
 - ecg
 - student-teacher
 - echocardiograms
+- medical
+pipeline_tag: other
 ---
-# EchoingECG
+
+# EchoingECG: An Electrocardiogram Cross-Modal Model for Echocardiogram Tasks
+
+The model was presented in the paper [EchoingECG: An Electrocardiogram Cross-Modal Model for Echocardiogram Tasks](https://huggingface.co/papers/2509.25791).
 
 EchoingECG is a probabilistic student-teacher model designed to improve cardiac function prediction from electrocardiograms (ECGs) by distilling knowledge from echocardiograms (ECHO). This approach leverages uncertainty-aware ECG embeddings and ECHO supervision, integrating Probabilistic Cross-Modal Embeddings (PCME++) and ECHO-CLIP, a vision-language pretrained model, to transfer ECHO knowledge into ECG representations.
 
-Please refer to our github for use: https://github.com/mcintoshML/EchoingECG
+You can find the official code and further details on our [GitHub repository](https://github.com/mcintoshML/EchoingECG).
 
-![EchoingECG Overview](assets/fig1_overview.png)
+## Features
+- ECHO-CLIP knowledge distillation
+- Probabilistic contrastive learning with PCME++
+- Outperforms state-of-the-art ECG models for ECHO prediction
+
+![EchoingECG Overview](https://huggingface.co/mcintoshML/EchoingECG/resolve/main/assets/fig1_overview.png)
 
 ## Installation
 Clone the repository and install dependencies:
@@ -23,6 +33,62 @@ cd EchoingECG
 pip install -r requirements.txt
 ```
 
+## Quick Start: Run EchoingECG in Jupyter Notebook
+Below is an example workflow using the provided demo notebook:
+
+```python
+import sys
+import yaml
+import torch
+from src.model.echoingecg_model import EchoingECG
+
+# Load model config
+with open("src/configs/model.yaml") as f:
+    model_cfg = yaml.safe_load(f)
+model = EchoingECG(model_cfg)
+model_weights = torch.load("echoingecg.pt", weights_only=True, map_location="cpu")
+model.load_state_dict(model_weights)
+
+# Example ECG input
+dummy_ecg = torch.zeros((1, 12, 1000)) # 10 seconds at 100Hz, 12 leads
+input = {"ecg": dummy_ecg}
+output = model(input)
+print(output["ecg"].keys()) # 'mean' and 'std' (probabilistic)
+print(output["ecg"]["mean"].shape, output["ecg"]["std"].shape)
+
+# Example text input
+from transformers import AutoTokenizer
+text_example = "ecg is normal"
+tokenizer = AutoTokenizer.from_pretrained("dmis-lab/biobert-v1.1", return_pt=True)
+tok_dict = tokenizer(text_example)
+input_model = {
+    "text": torch.tensor(tok_dict["input_ids"]).unsqueeze(0),
+    "attention_mask": torch.tensor(tok_dict["attention_mask"]).unsqueeze(0)
+}
+output = model(input_model)
+print(output["text"].keys()) # 'mean' and 'std'
+print(output["text"]["mean"].shape, output["text"]["std"].shape)
+
+# Load and scale an ECG properly
+from src.datasets.helpers import scale_ecg
+import joblib
+import numpy as np
+sc = joblib.load("ecg_scaler.pkl")
+_center = torch.from_numpy(sc.mean_.astype(np.float32))
+_scale = torch.from_numpy(sc.scale_.astype(np.float32)).clamp_min(1e-8)
+dummy_ecg = torch.zeros((1,12,1000))
+scaled_output = scale_ecg(_center, _scale, dummy_ecg)
+```
+
+## License
+This work is licensed under the **Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0)**.
+
+You may share this work for non-commercial purposes, with proper attribution, but you may not modify it or use it commercially.
+
+[![Creative Commons License](https://i.creativecommons.org/l/by-nc-nd/4.0/88x31.png)](https://creativecommons.org/licenses/by-nc-nd/4.0/)
+
+[View Full License Details](https://creativecommons.org/licenses/by-nc-nd/4.0/)
+
 ## Citation
 If you use EchoingECG in your research, please cite:
 ```