Room: ePoster Forums
Purpose: The tuning for the weighting factors in the objective function for an IMRT plan is driven by humans to achieve an ideal dose distribution. This is done manually in a tedious and time-consuming process based on trial and error and user experience. In the past fuzzy inference system (FIS) relying on static membership functions were used to tuning the weighting factors. The aim of this work is to propose a method for self-generating membership functions using an unsupervised learning method.
Methods: A dynamic membership function generator was implemented to translate linguistic humans tag (i.e. High or low dose) for different types of organs (target volume, the organ at risk and normal tissue) into a degree of truth. The membership function was generated using an iterative algorithm implemented in MATLAB using three mains points: central, left and right vertex for different types of membership functions, such as triangular, Gaussian, sigmoid and s-shaped. Then, they were evaluated in a Fuzzy logic guided inverse planning system to optimize the optimal combination of weighting factors in the objective function for an IMRT plan.
Results: The performance of the algorithm was examined using the C-Shape TG119 IMRT phantom using the variations of weighting factors and mean dose versus the iteration number as well as dose volume histograms. Fuzzy logic guided inverse planning system is capable of finding the optimal combination of mean dose and weighting factors for different anatomical structures involved in treatment planning within 20 iterations.
Conclusion: It is possible to use a feasible way to automatically tune the weighting factors for an IMRT plan under the guidance of FIS using an unsupervised membership functions generator without human intervention other than providing the treatment plan parameters and set of constraints.
Not Applicable / None Entered.