Optimal intermittent dose schedules for chemotherapy using genetic algorithm

• ALAM, Nadia ^[1] ; SULTANA, Munira ^[1] ; ALAM, M.S. ^[1] ; AL-MAMUN, M. A. ^[2] ; HOSSAIN, M. A. ^[2]
  1. [1] University of Dhaka

     University of Dhaka

     DCC (Kotwali), Bangladés

     
  2. [2] Computational Intelligence Group, Faculty of Engineering and Environment, University of Northumbria at Newcastle
• Localización: ADCAIJ: Advances in Distributed Computing and Artificial Intelligence Journal, ISSN-e 2255-2863, Vol. 2, Nº. 5, 2013, págs. 37-52
• Idioma: inglés
• Enlaces
  • Texto completo
• Resumen

  • In this paper, a design method for optimal cancer chemotherapy schedules via genetic algorithm (GA) is presented. The design targets the key objective of chemotherapy to minimize the size of cancer tumor after a predefined time with keeping toxic side effects in limit. This is a difficult target to achieve using conventional clinical methods due to poor therapeutic indices of existing anti-cancer drugs. Moreover, there are clinical limitations in treatment administration to maintain continuous treatment. Besides, carefully decided rest periods are recommended to for patient’s comfort. Three intermittent drug scheduling schemes are presented in this paper where GA is used to optimize the dose quantities and timings by satisfying several treatment constraints. All three schemes are found to be effective in total elimination of cancer tumor after an agreed treatment length. The number of cancer cells is found zero at the end of the treatment for all three cases with tolerable toxicity. Finally, two of the schemes, “Fixed interval variable dose (FIVD) and “Periodic dose” that are periodic in characteristic have been emphasized due to their additional simplicity in administration along with friendliness to patients. responses to the designed treatment schedules. Therefore the proposed design method is capable of planning effective, simple, patient friendly and acceptable chemotherapy schedules.