Each year 340,000 new cases of breast cancer are detected in the European Union. While there is much uncertainty about the causes of breast cancer, medical experts agree that if the disease is detected early the chances of survival are much higher. Therefore, screening programs have been established worldwide. Mammography is still the dominant modality in screening, despite its low sensitivity in dense breasts. In particular younger women would benefit from new screening techniques, such as ultrasound (US) which allows much better imaging in dense breast tissue, or MRI, capable of imaging functional parameters, like changes due to angiogenesis. Tomosynthesis is also increasingly used as an add-on or alternative to mammography and breast CT is also being investigated.
Our research focuses on developing automated image analysis methods aimed at helping radiologists with detecting and diagnosing breast cancer. Computer aided reading of mammograms is currently the largest application of CAD in practice. In the United States, most mammograms are nowadays read with CAD. Therefore, substantial improvement of automated detection of cancers in mammograms will have a large impact. DIAG has large annotated databases available and state of art algorithms that serve as a basis for further development. We are also closely involved in developing novel diagnostic breast imaging approaches, such as MRI, whole breast ultrasound, tomosynthesis and breast CT. Combining information obtained from multiple imaging modalities is also a challenging field of research which we are pursuing.