Optimization of Drug Scheduling in Cancer Chemotherapy by Using Orthogonal Genetic Algorithm

• Main Authors: Cheng-Chin Wu, 吳承晉
• Other Authors: Jinn-Tsong Tsai
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/91614986788608254927

碩士 === 國立屏東教育大學 === 資訊科學系 === 96 === The drug scheduling models are multimodal optimization problems, and their feasible solution spaces consist of several discontinuous subregions. This paper shows several better solving methods by using the orthogonal genetic algorithm. Throughout the entire cancer chemotherapy, the drug scheduling is not only important but also necessary. Doses of medicine prescribed by doctors vary for each patient depending on both their immunity and strength to the toxicity of the medicine. There have been several related models through mathematics methods about the optimal results of drug scheduling. This paper introduces Liang’s newly improved drug scheduling models in which they corrected several points that seemed unrealistic in Martin’s original research. Additionally, this paper focuses on the principle of drug scheduling－Cycle-wise Variable Representation, which will be reinterpreted. The orthogonal genetic algorithm which combined the Taguchi method of robust design for example, the application of orthogonal table experiment and signal-to-noise ratio, are very suitable to apply for the optimal results of drug scheduling models in complicated issues such as cancer chemotherapy. Meanwhile, this paper also proceeds to correlative studies about drug scheduling cycles and differences medicine provided doses of cancer chemotherapy, and use orthogonal genetic algorithm as the optimal method to analyze all feasibilities of parameters. The orthogonal genetic algorithm has been proven to be simpler and better astringent and stability.