Update README.md

README.md

@@ -1,3 +1,138 @@
----
-license: mit
----
+# 🧠 Text-Conditioned Latent Diffusion for Contrast-Enhanced CT Synthesis
+
+**Model Name**: `TUMSyn/ct-noncontrast-to-contrast`  
+**Model Type**: Fine-tuned `Stable Diffusion v1.5` for medical image-to-image translation  
+**Paper**: _Text-Conditioned Latent Diffusion Model for Synthesis of Contrast-Enhanced CT from Non-Contrast CT_  
+**Conference**: AAPM 2025 (Oral)  
+**Authors**: Mingjie Li, Yizheng Chen, Lei Xing, Michael Gensheimer  
+**Affiliation**: Stanford Radiation Oncology Department
+
+---
+
+## 🧬 Model Description
+
+This model is a fine-tuned version of **Stable Diffusion v1.5**, specialized for converting **non-contrast CT images** into **contrast-enhanced CT images**, guided by **textual phase prompts** (e.g., *venous phase*, *arterial phase*). It utilizes the `InstructPix2Pix` framework to enable flexible prompt-conditioned generation, enabling control over contrast timing without requiring explicit paired data.
+
+---
+
+## 💡 Key Features
+
+- 🧾 **Text-guided control** over contrast phase (arterial vs. venous)
+- 🖼️ Processes **2D CT slices** in image format (converted from DICOM)
+- 🏥 Focused on **clinical realism and anatomical fidelity**
+- 🧠 Reconstructs full 3D volume with NIfTI output support
+- ✅ Evaluated and presented as **Oral at AAPM 2025**
+
+---
+
+## 🛠️ Usage
+
+### 🔧 Requirements
+```bash
+pip install diffusers==0.25.0 nibabel pydicom tqdm pillow
+```
+
+### 📦 Load the Model
+```python
+from diffusers import StableDiffusionInstructPix2PixPipeline
+import torch
+
+pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(
+    "TUMSyn/ct-noncontrast-to-contrast", torch_dtype=torch.float16
+).to("cuda")
+generator = torch.Generator("cuda").manual_seed(0)
+```
+
+### 📝 Example Prompts
+
+- **Arterial Phase**
+  ```
+  Convert this non-contrast CT slice to mimic an arterial-phase contrast-enhanced CT.
+  Brighten and enhance the aorta, major arteries, and adjacent organ boundaries.
+  ```
+
+- **Venous Phase**
+  ```
+  Convert this non-contrast CT slice to mimic a venous-phase contrast-enhanced CT.
+  Brighten and enhance the portal and hepatic veins and emphasize organ boundaries.
+  ```
+
+### 🧪 Full Pipeline Example
+
+```python
+import os
+import numpy as np
+import nibabel as nib
+from PIL import Image
+from glob import glob
+from tqdm import tqdm
+from pydicom import dcmread
+from diffusers import StableDiffusionInstructPix2PixPipeline
+
+pipe = StableDiffusionInstructPix2PixPipeline.from_pretrained(
+    "TUMSyn/ct-noncontrast-to-contrast", torch_dtype=torch.float16
+).to("cuda")
+generator = torch.Generator("cuda").manual_seed(0)
+
+prompt = "Convert this non-contrast CT slice to mimic a venous-phase contrast-enhanced CT. Brighten and enhance the portal and hepatic veins, and emphasize organ boundaries."
+
+def load_dicom_folder(dicom_folder):
+    dicom_folder = os.path.join(dicom_folder, 'DICOM')
+    files = sorted(glob(os.path.join(dicom_folder, "*")))
+    slices = [dcmread(f).pixel_array.astype(np.float32) for f in files]
+    volume = np.stack(slices, axis=0)
+    volume += dcmread(files[0]).RescaleIntercept
+    volume = np.clip(volume, -1000, 1000)
+    return (volume + 1000) / 2000.0
+
+def process(volume):
+    results = []
+    for i in tqdm(range(volume.shape[0])):
+        img = Image.fromarray((volume[i] * 255).astype(np.uint8)).convert("RGB")
+        out = pipe(prompt, image=img, num_inference_steps=20,
+                   image_guidance_scale=1.5, guidance_scale=10,
+                   generator=generator).images[0]
+        gray = np.array(out.convert("L")).astype(np.float32) / 255.0
+        gray = gray * 2000 - 1000
+        results.append(gray)
+    return np.stack(results, axis=0)
+
+def save_nifti(volume, path):
+    nib.save(nib.Nifti1Image(volume, np.eye(4)), path)
+
+input_path = "/path/to/dicom_folder"
+output_path = "/path/to/output.nii.gz"
+
+vol = load_dicom_folder(input_path)
+out_vol = process(vol)
+save_nifti(out_vol, output_path)
+```
+
+---
+
+## 🧠 Intended Use
+
+- Medical research and simulation
+- Data augmentation for contrast-enhanced imaging
+- Exploratory analysis in non-contrast → contrast CT enhancement
+
+> ⚠️ **Disclaimer**: This model is for research purposes only. It is not intended for clinical decision-making or diagnostic use.
+
+---
+
+## 📝 Citation
+
+```
+@inproceedings{li2025text,
+  title={Text-Conditioned Latent Diffusion Model for Synthesis of Contrast-Enhanced CT from Non-Contrast CT},
+  author={Li, Mingjie and Chen, Yizheng and Xing, Lei and Gensheimer, Michael},
+  booktitle={AAPM Annual Meeting (Oral)},
+  year={2025}
+}
+```
+
+---
+
+## 🧾 License
+
+This model is released for **non-commercial research purposes only**. Please contact the authors if you wish to use it in clinical or commercial settings.