MILL (stands for Medical Image Labelling Laboratory) is an ideal platform and toolchain for labelling medical images.
Read this in other languages: 简体中文.
Head over to our release page and download the distribution suited to your machine. You'll get a native executable out of the box.
- Label medical images with free drawing
- Automatically close annotation curves and form marked areas
- Edit (or delete) each ROI independently with ease
- Export to binarized mask or JSON annotation code with a single button click
-
Run the executable file in the release package, it will automatically open the browser and enter the labelling interface.
-
Click on the "Choose File" button above to select desired medical image (PNG, JPG or JPEG format) or JSON annotation code. If it is JSON code, it will automatically load the medical image and its annotations and enter the "drawing mode".
-
Click and drag the mouse to start labelling. Mark one area each time. When the mouse is raised, the drawing path is automatically closed and filled to form a marked area. You can click the "Undo" button below to delete the marked area, or click the mode switch to switch to "Read-only mode" to adjust all current annotations.
-
After the annotation is completed, click “Download Annotation Map” to download the binarized annotation map generated by the label area (JPEG format). Click “Download Annotation Code” to download the JSON annotation code for re-labeling.