الفهرس | Only 14 pages are availabe for public view |
Abstract Cancer is one of the leading causes of morbidity and mortality worldwide. It is a highly complex group of diseases, involving uncontrolled abnormal cells growth; namely tumors. These cancer tumors can potentially infect the normal tissues of vital organs, e.g. the liver, kidney, and lung. One among the foremost common treatments of cancer is chemotherapy that aims to kill abnormal cells; but also, organs of the patients and normal cells are affected. In practice, it is typically tough to keep up optimum chemotherapy doses that may maximize the abnormal cell killing additionally as reducing side effects. To make the chemotherapeutical treatment effective, optimum drug planning is needed to balance between the helpful and poisonous effects of the cancer drugs. Standard clinical strategies often fail to search out the successful drug doses because of their inherent conflicting nature. The main objective of this thesis is to design a closed loop control system that is able to achieve the effective treatment with the least dose of chemotherapy by merging expert knowledge into automated closed-loop control of chemotherapy drug infusion. This control system must achieve safety elements for cancer patient such as toxicity limitation in the body and maximum limit for the chemotherapy dose. This thesis presents different control schemes for regulating intravenous chemotherapy drug delivery, based on intelligent controller with optimization algorithms. Using the mathematical model for cancer chemotherapy drug scheduling, the developed controllers constitutes the following two contributions: First, leveraging objectives of cancer tumor treatment are accomplished by combining the advantages of advanced closed-loop chemotherapy infusion control and optimization in a new framework. Second, clinical constraints; namely maximum allowed levels of both drug infusion dose and drug toxicity are implied in the design of optimized adaptive controller to ensure the medical safety of cancer patients. Fuzzy logic controller (FLC) is employed to merge expert physician knowledge into automated closed-loop control scheme of intravenous chemotherapy drug infusion. Furthermore, this drug infusion controller has been implemented using a microcontroller and tested in the hardware-in-the loop simulation framework to verify its robust performance. Optimal intuitionistic fuzzy logic control (IFLC) of chemotherapy drug delivery system is then presented in this thesis using two different optimization methods (Invasive Weed Optimization (IWO) and Ant Colony Optimization (ACO)). These optimization algorithms are utilized to optimize the intuitionistic fuzzy input-output scaling factors of the drug infusion system. The controller is designed and tested for tracking two clinically relevant chemotherapy dose scheduling protocols. Moreover, the patient safety is implied in the design of controller by considering maximum allowed level constraints on both drug infusion dose and drug-dose toxicity. Furthermore, computational results and comparative evaluations showed that the developed intuitionistic fuzzy controller outperforms other controllers in previous related studies to effectively minimize the tumor size at the end of cancer treatment time. Finally, this study showed that the patient sensitivity to toxicity has a great impact on the performance of cancer treatment and a suitable reference input scheduled within each cycle of treatment period has to be considered carefully. |