Projects & Software

Current research projects

ASSET: Automatic Fetal Biometry in First Trimester 3D Volumetric Ultrasound

AI-supported sonographic tomography for first-trimester fetal examination

MSKAssist: AI-assisted assessment of hand and wrist radiographs

Comprehensive analysis of hand and wrist radiographs with the help of artificial intelligence

Text mining to extract lifestyle classifications for prevention-related research

Extracting lifestyle information for prevention-related research from clinical notes using text mining and NLP techniques

MHub/IDC/Grand Challenge algorithm exchange

A collaboration between Radboudumc, MHub, and Imaging Data Commons (IDC), aimed at exchanging algorithms and investigating interoperability regarding exchange of data and algorithms.

PARADIGM: The diagnostic potential of 3D ultrasound with AI in maternity care

Determining the diagnostic potential of 3D ultrasound images acquired in the first trimester of pregnancy and analyzed with Artificial Intelligence

Quantifying IHC pathology stains

For the Pathology department we are developing an application that can unmix different IHC stains to aid analysist in quantifying pathology scans.

Coronary artery segmentation

For the department of Cardiology we are developing an application for the segmentation of coronary arteries in cardiac OCT recordings.

Adipose and muscular tissue segmentation

For the department of Health Evidence we are developing an application for the segmentation and quantification of adipose and muscular tissue in non-contrast abdominal CT.

In the context of the “Nationale Wetenschapsagenda” (NWA) roadmap for Value Creation through Responsible Access to and use of Big Data (VWdata) we are collecting data and setting up a challenge for the identification of healthy or abnormal MRI scans of the brain.

ICH and PHE segmentation

For the CONTRAST consortium and Dutch ICH Surgery Trial (DIST) we are developing an application for the segmentation and quantification of intracerebral hemorrhage and peri-hematomal edema in non-contrast CT.

Knee segmentation

For the Orthopedics department we are developing an application for the segmentation of the tibia, patella and fibula in non-contrast CT images of the legs.

Muscle ultrasound classification

For the Neurophysiology department we have developed an algorithm for the identification of abnormal muscle tissue in ultrasound images of the tibialis anterior.

ExaMode

The aim of ExaMode is to collect training data with limited human interaction for the processing of exascale volumes of healthcare data.

AI for health

A collaboration between Radboud University and Radboudumc aimed at developing innovations based on artificial intelligence that solve clinical problems.

AMI

Automation in Medical Imaging - a joint project between DIAG and Fraunhofer MEVIS aimed at the development of a generic platform for automatic medical image analysis.