For questions you can contact the press officers
- Annette Kik
tel. +31 (0)6 5157 4891
- Karin Blankers
tel. +31 (0)6 2422 1105
The Foundation Children Cancer free (Stichting Kinderen Kankervrij), the Department of Radiation Oncology of the Academic Medical Center (AMC) and the research group Life Sciences of Centrum Wiskunde & Informatica (CWI) join forces in a new research project to survey radiation-related long-term effects more accurately than ever before. The study is based on the 3D-reconstruction of the radiation dose of former patients whose long-term effects are known. The research lays the foundation for the development of improved radiation treatments and better decision support for the treatment of children with cancer.
It is for the first time that researchers focus this closely on survivors of childhood cancer that have been treated successfully more than twenty years ago. At a later stage, a large percentage (75%) of this group is confronted with one or more adverse effects of the treatment they underwent such as radiation-therapy related cancer, heart or lung problems.
To analyse in detail the relationship between these effects and the radiation dose that former patients have received, a 3D dose distribution is required that indicates the exact location and quantity of the delivered dose. However, of these former patients only 2D image information (X-rays) is available which lacks crucial information to calculate the 3D dose distribution.
To solve this challenging problem, the researchers will match data from patients that have been treated in the past, with patients that have been treated more recently and of whom a 3D CT scan is available. Based on this match the former radiation plan is reconstructed in 3D after correcting the CT scan of the patient for small anatomical variations. This reconstruction enables the researchers to study the relationship between the radiation dose and the long-term effects in detail.
Researchers from CWI will develop the underlying learning strategy to match the patients. For this purpose, state-of-the-art machine-learning and optimization algorithms will be developed and deployed. The first results from the study are expected to be available in 2016.