MedSeg
We ❤️ innovation in radiology.
MedSeg is a free online segmentation tool with AI.
Why MedSeg
MedSeg allows free and simple volume segmentation of organs, tissue and pathologies in radiological images. You can segment the images manually, or let our AI models do it for you.
Web-based
No installation needed. Runs locally, so patient data does not need to leave the computer. Works best with Google Chrome on a GPU-powered computer.
User-friendly
Easy-to-use tool for segmentation of radiology images with modern AI-based segmentation algorithms. Gaming and radiology-inspired controls.
Fast
Written in "vanilla JavaScript" for speed optimization. GPU hardware acceleration through Canvas and WebGL.
Launch MedSeg
Other Services
Segmentation service
We offer request-based radiologist-validated segmentations of CT or MRI datasets for a price. With modern and custom AI models, we can annotate precisely and effectively. The result can be masks, volumes or 3D models/meshes of any file format.
Research
We are very open to research collaboration. To facilitate research in this field, we are creating open-access segmentations whenever we find the time.
Learn MoreTraining AI-models
You can contribute by letting us train models on your data. These models would be shared with everyone in MedSeg. Another way to contribute is to do segmentations and help us grow the open-access database.
Database
External databases
Finding good quality data of sufficient abundance is a well known problem in AI. Here you'll find our top choices with detailed instructions and download walkthroughs
Read More
COVID-19
Forum
About
Segmentation of radiological images is important in many fields. Volumetry, visualization including VR/AR, 3D printing, radiotherapy, (co-)registration, and many other post-processing tools are some examples that often rely on segmentation.
Unfortunately, segmentation is still generally regarded as laborious and uninteresting by many. However recent progress in the field of computer vision has enabled the use of AI for both automated and interactive segmentation. For radiology, we hope that these new developments will hasten broad clinical adoption of volumetry to replace the subjective ruler-based distance measurements that are still the current clinical standard.
MedSeg's vision
We believe that AI-based segmentation is technologically ready for broad clinical implementation. Although MedSeg is not meant for direct clinical use, we hope that it, together with our open-access segmentation database, will facilitate faster development of clinically validated models. Radiological segmentation should be simple, fun and accessible for everyone.
If you are a radiologist, we hope that you share our vision and would consider helping us with annotating more data (CT and MRI). This data is made freely available so that anyone can download and develop tools that we believe can enhance radiology in general. If you are interested, please get in touch with us. We can provide openly accessible data that can be used for this purpose.
MedSeg is based on the following openly available tools and resources:
DICOM reader: https://github.com/rii-mango/Daikon
Nifti reader: https://github.com/rii-mango/NIFTI-Reader-JS
Deep learning (DL) in browser: https://www.tensorflow.org/js
Development of DL models: https://keras.io/ + https://www.tensorflow.org/
DeepGrow module: https://arxiv.org/abs/1903.08205
Ideas and experience from a Python-based segmentation tool with AI-capabilities. RILContour: https://link.springer.com/article/10.1007/s10278-019-00232-0
Another great advanced application with many post-processing tools, 3D Slicer: https://www.slicer.org/
Overview of web-based DICOM viewers: https://medevel.com/14-best-browser-web-based-dicom-viewers-projects/