KOMET – Transfer platform in the field of medical technology: segmentation of soft tissue structures of the neck in MRI data

In otorhinolaryngology, the diagnosis and therapy of malignant tumor diseases in the area of the mouth, nose and jaw represents an important area. The type of therapy depends on the overall assessment of the disease. Special care must be taken when resecting tumors or lymph nodes with metastases, as damage to nearby functional structures, such as large cervical vessels and cranial nerves, can lead to a significant reduction in the patient’s quality of life. Imaging techniques such as ultrasound, CT, MRI, or PET can be used to determine the malignancy of lymph nodes, for example, based on their size. However, size measurement is quite nonspecific and can lead to false positive or false negative findings. More specific is the detection of necrosis using MR imaging.

For resection of malignant tissue, image analysis and visualization techniques have already been developed for evaluation of CT data and planning of tumor surgery. The aforementioned functional structures should be spared. These techniques are already applied in TumorTherapyManager.

In this project, the techniques developed for CT data will be tested on MRI data and, if necessary, adapted so that surgical planning is also possible on this data. The expansion of surgical planning is motivated by the fact that MRI provides better soft tissue contrast and thus infiltration of at-risk structures can be better estimated. However, difficulties, e.g., due to inhomogeneities, geometric distortions, or different intensity values, are to be expected when analyzing MRI data. The adapted or newly developed image analysis and visualization techniques will finally be clinically evaluated.



Sorry, no publications matched your criteria.