| High Intensity Focused Ultrasound (HIFU) has demonstrated its effectiveness in killing small tumors while preserving surrounding tissue. Phases I and II clinical trials have been carried out in the United States and elsewhere using HIFU to treat different tissue and diseases, including benign prostate hyperplasia (BPH), prostate, bladder, and kidney tumors. When treating a larger volume, however, multiple foci are needed to cover it. Simply scanning the focused beam across the volume with a fixed power output level would either result in overheating of the intervening tissue or require minutes of cooling time between successive lesion formations, resulting in extraordinarily long treatment time. Thus, a better way of depositing the acoustic power in space and time is necessary, which is the task of thermal dose optimization. Monitoring of the treatment process using ultrasound or other imaging modalities is also indispensable for consistent localization and complete ablation.; In this dissertation the issue of thermal dose optimization is addressed. First, techniques for producing an acoustic field pattern via both single focus formation and multiple focusing are reviewed and practical recommendations made. Then a two-step thermal dose optimization technique is developed. Previously-developed methods require brute force searches which are very time consuming while not able to create an optimal result under a general situation. The thermal dose optimization algorithm developed here consists of (1) a closed-form solution to the bioheat transfer equation (BHTE); (2) a Gaussian model for parameterizing temperature rise and acoustic intensity patterns; and (3) a two-step optimization technique to obtain model parameters optimizing the thermal dose distribution while meeting given constraints.; Examples are given to demonstrate the effectiveness of the algorithm and its robustness under different initial conditions and under different target region sizes and numbers of foci. The thermal dose optimization algorithm developed in this work shows promise as an efficient and effective tool for treatment planning in ultrasound tissue ablation. Combined with on-line monitoring and feedback of the treatment process, this method has the potential to bring treatment of larger target volumes closer to reality. |
