In radiation therapy for thoracic cancer treatments it is crucial to keep the doses received by the volumes of the lungs and heart as low as possible. Due to high dose conformation, Pencil Beam Scanned (PBS) proton therapy holds the potential to signicantly reduce doses to surrounding healthy tissue. For lung or liver cancer, the delivered beams are steered to follow tumour motion ensuring the best combination of target coverage and dose conformation in the presence of respiratory motion. Successful tumour tracking, however, requires accurate methods for determining both tumour position and motion in real-time. As such, ultrasound (US) and optical surface imaging are interesting modalities as they are non-invasive and the patient is not exposed to additional radiation dose. Unfortunately, neither of the two techniques is able to image lung tumours directly.

It is the aim of this project to develop methods by which US and surface imaging of the upper abdomen are used to predict three-dimensional motions in the lung. We develop an indirect prediction scheme where surrogate structures in the liver, on the diaphragm and abdominal wall in combination with a patient-specic motion model of the lung are used for prediction.

This PhD thesis is embedded in a project founded by the Swiss National Science Foundation (SNSF) and part of a cooperation between the Centre for Proton Therapy at Paul Scherrer Institute (PSI), the Radiological Physics group at the University Hospital Basel, the Department of Radiation Oncology at University Hospital Zurich, and the University Hospitals of Geneva.

Project leader: Alina Giger