Research

Imaging is a cornerstone of modern medicine. The amount of imaging that is performed is growing, the number of modalities is growing, and the resolution and dimensionality of the scans is increasing. Our research focuses on creating software to let computers help physicians in the image interpretation process. Our activities cover the whole spectrum from research, development, evaluation, to translation to the clinic. We therefore work closely with clinicians and have a disease oriented approach. We develop image processing and machine learning algorithms, but also carry out observer studies, and develop and support software that runs in a clinical environment.

Below you can find more information about the various areas we work in.

4DCT

Computed Tomography (CT) imaging of the brain is an important step for diagnosis and treatment planning. The latest advances in CT technology enable dynamic scanning or perfusion imaging covering the complete brain with a temporal resolution of 0.35 seconds. This means you can see the in-flow and out-flow of the blood in real-time and obtain detailed functional information in addition to the anatomical information of the brain More...

White Matter Lesions

White matter lesions (WMLs) are areas of demyelinated cells found in the white matter of the brain. Minor cases that are commonly found in people over 65 years old, are thought to be the result of normal aging. However, there are other factors that contribute to the presence and amount of WMLs such as migraine headaches, stroke, or progressive neurological diseases that cause brain degeneration; such as multiple sclerosis and Alzheimer’s. More...

Retinal Image Analysis

The availability of cheap cameras to take direct images of the retina, fundus photography, makes it possible to scan the eye for several eye diseases with a simple, non-invasive method. Within DIAG, research has focused on automatic early detection of diabetic retinopathy from fundus photographs. We are currently also applying the computer-aided detection and quantification techniques we have developed to diagnosis and quantification of macular degeneration. More...

Breast cancer CAD

The earlier breast cancer is found, the more likely it can be successfully treated. We have found evidence that interpretation failure is a more common cause of missing cancers in screening than perceptual oversights. By detecting these missed cancers earlier the effect of screening can be increased. Computer aided detection (CAD) methods have been introduced as a way to avoid perception errors. More...

Breast density and cancer risk

Mammographic breast density, which reflects the amount of radiographically dense tissue on a mammogram, has been shown to be one of the strongest risk factor for breast cancer. Numerous studies have reported that in women with high breast density the risk of breast cancer is increased 2 to 6 fold compared to women with low breast density. we have developed a fully automated method that integrates and extends state of the art breast density segmentation techniques. More...

Breast cancer detection in automated breast ultrasound risk

As a complementary modality to mammography, breast ultrasound has a high specificity and its sensitivity can surpass that of mammography for patients with highly dense breast. The aim of this project is to develop a computer-aided detection (CAD) system to play an important role in the future, as it has the potential to make reading more efficient and reduce reading errors. More...

Lung Cancer

Cancer is a leading cause of death worldwide and accounted for 7.6 million deaths (around 13% of all deaths) in 2008. Lung cancer is the most common cause of cancer-related death in men and women and accounted for 1.4 million deaths in 2008 and accounts for more annual deaths than breast, colon and prostate cancers combined. In this project, we focus on extending and improving our current nodule CAD algorithms in close collaboration with MeVis Medical Solutions AG. More...

Computer-Aided Detection of Tuberculosis (CAD4TB)

Tuberculosis (TB) is one of the most deadly diseases on earth. Every year around nine million people get TB and almost two million die from it. Together with Delft Diagnostic Imaging we developed a prototype computer-aided detection system for TB in 2001. Over ten digital units are currently operational within the CAD4TB project, in Zambia, South Africa, Gambia, Tanzania, Ghana. More countries are expected to follow.. More...

Chest CT analysis

Research in DIAG focuses on the analysis of chest x-ray and chest CT. For x-ray we are developing tools to detect tuberculosis and lung cancer, work on algorithms to remove normal anatomical structures from the radiographs and to detect changes between baseline and follow-up exams. For CT we are working on computer-aided detection of nodules, which can be small lung tumors, and we have developed a toolkit for analysis of the lungs, the lobes, airways and vessel trees. More...

CAD Assessment

Several studies have reported on the potential of CAD, but CAD has not yet proven to be beneficial in clinical practice. Performance depends on the readers capabilities, sensitivity and false positive rate of the CAD system, and the ability of human readers to differentiate between true positive (TP) and false positive (FP) CAD markers. We intend to perform a number of observer studies aimed at measuring the clinical benefit of CAD and investigating the effect of different ways to use CAD. More...

Computer-aided detection of prostate cancer

Prostate cancer is the second most common cancer in men, with 10000 new cases and 2500 deaths every year in the Netherlands Dutch Cancer Society. Over the past decade prostate MRI has shown to be a great candidate for prostate cancer detection. Computer algorithms allow us to condense all the information coming from the scanner into images with a high information content. More...

Computer-aided detection of metastatic pelvic lymph nodes

The detection of nodal metastases is of utmost importance for prognosis and choice of treatment in prostate cancer. A metastasis in one lymph node turns prostate cancer from a local to a systemic disease, with fewer curative options. The interpretation time of MRL can be reduced and the accuracy further improved by using a Computer-Aided Detection (CAD) system that detects the lymph nodes in the MRL images and subsequently determines which of them contain metastases. More...

MR guided ultrasound prostate biopsy

Prostate biopsies are performed using systematic transrectal ultrasound (TRUS). However, the low sensitivity and low detection rate after repeated biopsies make it less than ideal. Prostate MR, on the other hand, has been demonstrated to accurately localize the cancer in the prostate. However, fully MR guided biopsies are costly and complex. We aim to combine the advantages of both techniques. More...