Update README.md

README.md

@@ -1,3 +1,44 @@
----
-license: cc-by-nc-4.0
----
+---
+license: cc-by-nc-4.0
+---
+### CT-Scroll: A Global-Local Attention Model for 3D Chest CT Volumes 🩺👨🏻‍⚕️
+Official hub for the paper "Imitating Radiological Scrolling: A Global-Local Attention Model for 3D Chest CT Volumes Multi-Label Anomaly Classification".
+
+Paper accepted for [MIDL](https://www.midl.io/) 2025 : [arXiv submission](https://arxiv.org/abs/2503.20652)
+
+## Method Overview
+
+The CT-Scroll architecture consists of three main components. (1) Axial slices of the volume are grouped into triplets and processed by a ResNet followed by a GAP layer, producing a vector representation per triplet. (2) The Scrolling Block then refines these embedded visual tokens using both global and local attention mechanisms. (3) Finally, the aggregated representations are fed into a classification head to predict anomalies.
+
+<img src="./figures/method_overview.jpg" width="1000">
+
+## Clone the Repository
+
+To clone this repository, use the following command:
+
+```bash
+git clone https://huggingface.co/theodp/ct-scroll
+```
+
+## Code repository
+
+The project source code can be found [here](https://github.com/theodpzz/ct-agrg).
+
+## Acknowledgments
+
+We acknowledge [Hamamci et al. 2024](http://arxiv.org/abs/2403.17834) for making the [CT-RATE dataset](https://huggingface.co/datasets/ibrahimhamamci/CT-RATE) available, and [Draelos et al. 2021](https://www.sciencedirect.com/science/article/pii/S1361841520302218) for making the [Rad-ChestCT dataset](https://zenodo.org/records/6406114) available. We would like to express our gratitude to [CT-CLIP GitHub repository](https://github.com/ibrahimethemhamamci/CT-CLIP) for providing a codebase that served as a foundation for this project.
+
+## Citation
+
+> [!IMPORTANT]  
+> This project is based on the work by Di Piazza et al. If you use this code in your research, please cite the following paper:
+
+```BibTeX
+@inproceedings{dpzz2025ctscroll,
+  author    = {Di Piazza Theo, Carole Lazarus, Olivier Nempont and Loic Boussel},
+  title     = {Imitating Radiological Scrolling: A Global-Local Attention Model for 3D Chest CT Volumes Multi-Label Anomaly Classification},
+  booktitle = {Proceedings of Medical Imaging with Deep Learning (MIDL)},
+  year      = {2025},
+  organization = {MIDL},
+}
+```