- Project description
- Scientific background
- Project partners
Goals
Kohala is a project supported by the „Wirtschaftsförderungsgesellschaft HA Hessen Agentur GmbH“. The project aims at developing a novel system for automated segmentation of organs and structures within the head and neck region which is going to be used in radiation therapy.
Vision
In contrast to existing approaches, a new method is being developed and tested which aims at increasing the success rate of the automatic segmentation significantly and thereby reducing the usual manual post-processing on the individual patient data set by a clinician which is very time consuming. To achieve this goal we use the new concept, called "Articulated Atlas", which has been developed and demonstrated at the Fraunhofer IGD.
Methods
In order to perform this fully automatic segmentation on individual patient data sets an “Articulated Atlas” model is trained from a set of training data first.
Here, the shapes and sizes of the individual structures (bones, soft tissue and tubular structures) are learned. Therefore, statistical shape models are used, which contain an average model of the structure as well as 95% of the seen training data sets. That means the model is highly variable but on the other hand limits the structures to cover only natural deformations.
In addition, the positions of the structures to each other are learned. This ensures that all of the structures in the final segmentation are located correctly and prevents that any structure is completely misplaced.
Finally, the “Articulated atlas” is transferred to an individual patient data set to perform an automatic segmentation. Here, the previously trained knowledge of the shapes and sizes of the individual structures and the relative positions to one another is used to achieve an accurate fully automatic segmentation.
Consortium
The consortium consists of four partners:
- Fraunhofer IGD (software development)
- MedCom GmbH (software development, integration)
- Klinikum Offenbach (data acquisition, validation)
- Universitätsklinikum Gießen - Marburg (data acquisition, validation)