42 lines
1.7 KiB
Markdown
42 lines
1.7 KiB
Markdown
<div align="center">
|
|
<img width="1200" height="475" alt="GHBanner" src="https://github.com/user-attachments/assets/0aa67016-6eaf-458a-adb2-6e31a0763ed6" />
|
|
</div>
|
|
|
|
# Run and deploy your AI Studio app
|
|
|
|
This contains everything you need to run your app locally.
|
|
|
|
View your app in AI Studio: https://ai.studio/apps/b9650cf9-3b26-4e84-a699-78ba380bb4db
|
|
|
|
## Run Locally
|
|
|
|
**Prerequisites:** Node.js, Python 3.8+
|
|
|
|
1. Install Python dependencies in the project root:
|
|
`pip install -r ../requirements.txt`
|
|
2. Install website dependencies:
|
|
`npm install`
|
|
3. Start the local Python backend:
|
|
`npm run backend`
|
|
4. In another terminal, run the website:
|
|
`npm run dev`
|
|
|
|
The website talks to `http://127.0.0.1:8787` and maps UI actions to:
|
|
|
|
- `head_extension_app.py`: preview and four-state DICOM deformation output
|
|
- `generate_head_extension_video.py`: 0° to target-angle MP4 generation
|
|
- `video_generator_app.py`: kept as the desktop GUI wrapper for the same video generator
|
|
|
|
## Data Flow
|
|
|
|
The Image Library is now the source of DICOM data for the workstation:
|
|
|
|
1. Open `数据影像库`.
|
|
2. Click `上传文件夹` to choose a folder that contains `.dcm` files, or click `上传压缩包` to upload a `.zip` archive containing `.dcm` files.
|
|
3. Click `调阅工作站` on a library item.
|
|
4. The `影像变换工作站` will run preview, four-state deformation, and video generation from that selected library dataset.
|
|
5. Four-state deformation results are packaged by the backend as a downloadable `.zip` file under `../web_results/`.
|
|
6. Generated videos are also written under `../web_results/` and exposed as downloadable `.mp4` files.
|
|
|
|
Uploaded DICOM datasets are stored by the local backend under `../web_library/`.
|