Head and Neck Cancer is the 7th most common cancer worldwide. In addition the reoccurrence probability is one of the highest among cancer. Until today the causes of the diseases reoccurrence are not well known.
The goal of the OraMod project is the development of a statistical model for the prediction of an oral cancers reoccurrence probability. The model is based on the biomarkers of 150 patients. The biomarkers consist of genetic biomarkers, biomarkers extracted from medical image data and meta-information of the patient (e.g. smoker, sex, age). Once the model has been successfully trained it can be used for the prediction of the survival rate and reoccurrence probability of future patients.
We at Fraunhofer IGD have developed a medical image software based on the MITK toolkit for our clinical project partners, VUMC from Amsterdam, the university clinic Parma and the university clinic Düsseldorf. The main goal of the OraMod Imaging Software is to enable the clinical partners in the OraMod project to enhance clinical images with segmentations of tumors and lymph nodes as well as semi-automatically classify the found medical conditions of the patient and to extract these biomarkers for later analysis.
The image data can be accessed directly from a clinical PACS server. The clinician’s task is to do a segmentation of the tumor and all relevant lymph nodes. This can be an extremely time consuming task, if done manually. Therefore, several algorithms have been developed to reduce the workload of the clinicians as much as possible.
Additionally the software features algorithms which can automatically determine the T-staging of tumor as well as the N-staging of the lymph nodes once the needed parameters are present.
The extracted biomarkers finally can be uploaded to a clinical PACs server and the OraMod research platform for evaluation purposes.
More information about the project can be found at www.oramod.eu