The VinDr Lab is a Data Platform for Medical AI that enables building high-quality datasets and algorithms with the lean process and advanced annotation features. The software is provided by the Medical Imaging team at Vingroup Big Data Institute (VinBigdata).
VinDr Lab is an open-source platform for medical image annotation. It has been developed to remove the ground-truth barrier AI teams met to build meaningful medical AI applications. VinDr Lab provides a high-level web interface equipped with advanced annotation tools and project management features.
The following features have been implemented:
- Manage full medical data cycle at they study level
- Track project progress and status of each task
- Create and share label groups
- Organize labels by order and hierarchy
- Annotation tools: Bounding box, Polygon, Brush, Notes, Comments
- Control versions of exported labels
This project is still in active development, so some options may not be fully implemented yet.
Future features
- Annotation in MPR view
- AI itergration
You can read more about the platform on vindr.ai/vindr-lab.
This repo is the "master" repo for all VinDr Lab-related projects. It hosts the documentation and other misc. resources for VinDr Lab. Code for other projects, like backend and frontend, is hosted in other repositories.
All VinDr Lab projects and repositories underneath the VinBigdata Medical GitHub organization are listed here. To learn more about each project, see the README in each repository.
Repository | Description | Maintainers |
---|---|---|
vinbigdata-medical/vindr-lab | Documentation and other general resources related to VinDr Lab | @lego1st, @fuzzysource |
vinbigdata-medical/vindr-lab-deployment | Getting started with VinDr Lab. We support Kubernetes or Docker deployments. | @iamatsundere |
vinbigdata-medical/vindr-lab-api | Middleware layer between the user interface and backend systems | @iamatsundere |
vinbigdata-medical/vindr-lab-id-generator | For each study or task of a project, this service generates a new integer value based on the key as a string. | @iamatsundere |
vinbigdata-medical/vindr-lab-uploader | Uploader modifies and transfers the DICOM files to the database. | @iamatsundere |
vinbigdata-medical/vindr-lab-viewer | Frontend: Viewer allows the manipulation, annotation, and serialization of observations and many more features. | @HoAnhVan |
vinbigdata-medical/vindr-lab-dashboard | Frontend: Dashboard support data management and labeling administration | @HoAnhVan |
Note: Another way to find all VinDrLab-related repositories is to go to Vinbigdata Medical GitHub main page and enter the search term "vindr-lab", like here.
Our project requires at least 4GB of RAM system for the best performance and experiment. We support both Kubernetes and Docker deployments. Please visit VinDr Lab Deployment to start !
Project documentation is hosted on our Github page. We are hoping to establish a more user-friendly version soon.
For quick questions, there's no need to open an issue as you can reach us on Gitter.
If you're an experienced dev and interested in contributing to VinDr Lab:
- Please read our Contribution Guidelines for details on what and how you can contribute.
- Check out our Development Project Board for Product Roadmap (coming soon).
- Look for tasks under the Issues tab in each repo.
- Join our Gitter channel.
To acknowledge VinDr Lab in an academic publication, please cite
@misc{VinDr Lab,
author = {Nghia T. Nguyen, Phuc T. Truong, Van T. Ho, Trung V. Nguyen, Hieu T. Pham, Mi T. Nguyen, Long T. Dam, Ha Q. Nguyen},
title = {{VinDr Lab: A Data Platform for Medical AI}},
howpublished = {\url{https://github.com/vinbigdata-medical/vindr-lab}},
year = {2021}
}
Note: If you use or find this repository helpful, please take the time to star this repository on Github. This is an easy way for us to assess the adoption and it can help us obtain future funding for the project.
This work is supported primarily by Vingroup Big Data Institute
Lego1st | fuzzysource | iamatsundere | HoAnhVan | trung1704ptit | EdwardPham1615 | minguyenn |