Algorithms

Here we have made available several machine learning algorithms for medical imaging. If you are granted access to an algorithm you will be able to try the algorithm out by uploading your own anonymised medical imaging data. Several web-based workstations are available for interacting with the results. Please contact us if you would like to make your own algorithm available here.

Spleen Segmentation Logo
Spleen Segmentation

This website reproduces the results obtained by the segmentation algorithm presented in the paper entitled "Fully automatic volume measurement of the spleen in CT using deep learning" and allows users to test this algorithm on new cases. The paper is currently under review. Every user is allowed to submit up to 5 thorax-abdomen CT scans per day. Limitations: This algorithm was trained using the following data settings: Contrast-enhanced thorax-abdomen CT scans, slice thickness up of 3mm or less. Different settings than the mentioned above might affect the performance of the algorithm.

Vertebra segmentation and labeling Logo
Vertebra segmentation and labeling

Tumor Detection in Lymph Nodes Logo
Tumor Detection in Lymph Nodes

Tissue-Background Segmentation Logo
Tissue-Background Segmentation

Gleason Grading of Prostate Biopsies Logo
Gleason Grading of Prostate Biopsies

Automated Gleason grading of prostate biopsies following the Gleason Grade Group system.

Calcium scoring Logo
Calcium scoring

Automatic detection of calcifications of the coronary arteries, the aorta and the aortic and mitral valves in CT scans.