Towards adaptive IMRT sequencing for the MR-linac. (6th March 2015)

• Record Type: Journal Article
• Authors: Kontaxis, C; Bol, G H; Lagendijk, J J W; Raaymakers, B W
• Is Part Of: Physics in medicine & biology. Volume 60:Number 6(2015:Mar.)
• Page Start: 2493
• Page End: 2509
• Publication Date: 2015-03-06
• Subjects: IMRT -- adaptive -- sequencer -- MR-linac; Biophysics -- Periodicals; Medical physics -- Periodicals; 610.153
• DOI: 10.1088/0031-9155/60/6/2493
• Languages: English
• ISSNs: 0031-9155

Abstract

Abstract: The MRI linear accelerator (MR-linac) that is currently being installed in the University Medical Center Utrecht (Utrecht, The Netherlands), will be able to track the patient's target(s) and Organ(s) At Risk during radiation delivery. In this paper, we present a treatment planning system for intensity-modulated radiotherapy (IMRT). It is capable of Adaptive Radiotherapy and consists of a GPU Monte Carlo dose engine, an inverse dose optimization algorithm and a novel adaptive sequencing algorithm. The system is able to compensate for patient anatomy changes and enables radiation delivery immediately from the first calculated segment. IMRT plans meeting all clinical constraints were generated for two breast cases, one spinal bone metastasis case, two prostate cases with integrated boost regions and one head and neck case. These plans were generated by the segment weighted version of our algorithm, in a 0 T environment in order to test the feasibility of the new sequencing strategy in current clinical conditions, yielding very small differences between the fluence and sequenced distributions. All plans went through stringent experimental quality assurance on Delta4 and passed all clinical tests currently performed in our institute. A new inter-fraction adaptation scheme built on top of this algorithm is also proposed that enables convergence to the ideal dose distribution without the need of a final segment weight optimization. The first results of this method confirm that convergence is achieved within the first fractions of the treatment. These features combined will lead to a fully adaptive intra-fraction planning system able to take into account patient anatomy updates during treatment.