Technology Portfolio

Here are some of our fields of expertise

Image Segmentation

Segmentation is a very common task in image analysis and consists in extracting the contours of a structure of interest in the image. The ImFusion Suite provides a set of tools to help the user segment any organ, from interactive algorithms to specialized automated workflows.

Interactive Segmentation

Our software features a powerful interactive segmentation that can be used to segment any structure in a 2D or 3D image. The user is asked to draw regions of the images that are inside and outside the structure of interest (see figure). At any time, the segmentation can be previewed and the user input can be refined. When the result is satisfactory, it can be exported either as a label map or as a mesh.

This algorithm can be combined with a machine learning model so that those regions are automatically predicted, turning this method into a fully automated pipeline.

A variety of post-processing methods on segmentations are available from the user interface of our software, including smoothing, decimating, cropping, merging, connected component analysis, etc.

Interactive segmentation of a 3D volume in our software. As the user paints inside (green) and outside (red) regions, the segmentation (yellow) is updated and can then be exported as a mesh or a label map.

The ImFusion Suite also features algorithms to automatically refine existing segmentations, for instance generated by other software tools or by means of manual annotation. In the figure herebelow, we show the refinement of a femur segmentation in a CT volume using image matting.

(Top Left) Original image. (Top Right) Comparison of the two segmentations, the old one is in red while the new one is in green. (Bottom Left) Mask corresponding to the original segmentation. (Bottom Right) Refined mask after image matting.

Automated Segmentation

Since each segmentation problem shows very specific characteristics (prior knowledge on the appearance, the shape, etc.), segmentations algorithms have to be tailored in order to be robust. We therefore also have a number of more dedicated algorithms in our portfolio, as exemplarily demonstrated in the figures below.

Robust and accurate bone segmentation from ultrasound sweeps

Automatic neck segmentation from CT volumes

Segmentation of a vein in a large 3D ultrasound volume