Multiobjective optimization technique for treatment planning in HIFU

In this study, a multiobjective optimization technique developed by the author is introduced to solve the scanning sequence (or switching) plan of multiple-focus patterns, satisfied with the objectives of uniform high dose to the block and fast ablation. In case of multiple focusing, the shape and scan time of the single block which is a fundamental factor for planning the positions of the blocks depends on the sequence plan. In the previous study, new HIFU transducer architecture with sparse array (some of total elements), which can generate multiple-focus patterns simultaneously was designed and evaluated. Planning the sequence of multiple-focus scan for the block is a time consuming and monotonous task, but requires at the same time expertise and precision. Furthermore, due to the complex characteristics of HIFU treatment, it is difficult to optimize manually. Besides, the objectives of uniform high dose to the block and fast ablation are invariably in conflict, often requiring compromises to be made between them when selecting the best sequence plan for a particular condition. The multiobjective optimization problem used as the objectives of uniform high dose to the block and fast ablation was formulated in term of the scanning sequence of multiple-focus patterns. Pareto-based continuous evolutionary algorithms for multiobjective optimization (MOPCEAs) developed by the author was used to solve the sequence planning. In multiobjective sequence planning, we produced diverse Pareto-optimal solutions, and the treatment planner could navigate through them intuitively.