Towards a predictive model for functional loss after oral cancer treatment

For oral cavity and oropharynx tumours the estimation of functional inoperability is a key factor in treatment planning. Functional inoperability means the tumour can be resected radically, but the expected functional loss is unacceptable, causing a different treatment, mostly chemoradiation, is proposed. Due to patient-specific factors the estimation of the functional results are uncertain. To reach objective and accurate predictions the project “Virtual Therapy for Head & Neck Cancer‿ was launched, which aims on the development of a patient-specific biomechanical model of the oral cavity and oropharynx, on which a virtual treatment can be performed, resulting in a visualization of the functional results. The individual characteristics make an accurate prediction of postoperative functionalities difficult. In the framework of this dissertation, studies are performed to obtain knowledge on variability in anatomical structures and innervation of musculature, and ways to implement them in a biomechanical model. Using finite element modelling realistic tongue motions as well as the effect of fibrotic tissue could be simulated. Furthermore, patient-friendly surface EMG contains sufficient information to predict positions and motions of the lips. In future research those signals can be used to drive the biomechanical model. Peroperative stimulation of the hypoglossal nerve and simultaneous measurements of tongue motions indicated that more anatomic variations of this nerve exist than initially thought. Even the bilateral branching pattern within the patient can differ. A setup to analyse the range of motion of the tongue was created and proven to be reliable. This setup can be used to develop a grading system in which patients can be categorized in risk levels for functional problems. This will also serve as an important feedback to evaluate the biomechanical model. The studies, described in this dissertation lead to methodologies, which can be performed clinically to acquire patient-specific data regarding muscle innervations and the mobility of the lips and tongue. Furthermore, more knowledge is obtained on the anatomical diversity, but also on the possibilities to incorporate patient-specific information in biomechanical models, with the ability to perform virtual treatments.