A measurement-based control design approach for efficient cancer chemotherapy

Sofiane Khadraoui, Fouzi Harrou, Hazem Nounou, Mohamed Nounou, Aniruddha Datta, Shankar P. Bhattacharyya

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

This paper presents a new measurement-based control design method for cancer chemotherapy. Cancer chemotherapy aims basically at simultaneously eradicating or significantly reducing the number of cancer cells and maintaining tolerable levels of drug concentration and toxicity. To achieve such aim, drugs are often injected into the patient's body according to a drug schedule specifying the drug dose and delivery time. Several strategies for planning cancer chemotherapy have been developed in the literature, where evolutionary algorithms have been applied to find optimal drug schedules of cancer treatment under constraints on some key treatment parameters such as drug concentration and toxic side effects. In such methods, the amount of drug doses, delivered in the body at each time during the treatment, does not depend on the current drug concentration, toxicity level, and/or number of cancer cells. Successful design of chemotherapy drug scheduling requires the availability of an accurate mathematical model that perfectly predicts the number of cancerous cells and describes effects of treatment. Several models with either complex or simple structures are available in the literature. Complex-structure models are proposed to deeply understand interactions between cancer and normal cells that affect the performance of the cancer chemotherapy. Nevertheless, such complex models are based on a high-order set of differential equations which can be difficult to solve. Simple-structure models, which are often obtained on the basis of some simplifying assumptions, can be viewed only as an approximation of the cancer system. Hence, designing chemotherapy drug schedules on the basis of simplified models may result in unsuccessful cancer treatment. Unlike conventional control strategies for cancer chemotherapy, our attempt in this paper is to address the problem of designing a control system for cancer treatment using a set of frequency-domain data. Hence, a two-degree-of-freedom PID (proportional-integral-derivative) control scheme is proposed to control cancer growth. These PID controllers are designed to simultaneously provide the optimal amount of drug doses to be delivered into the patient's body according to the current drug concentration and toxicity level, and maintain the drug concentration and toxicity levels within their pre-specified ranges. The proposed cancer control technique is validated through a first simulation example. Another example to control biological systems is also presented to show the feasibility of the proposed method. Simulation results obtained have demonstrated the capability of the proposed control scheme to address cancer chemotherapy problems.

Original languageEnglish
Pages (from-to)108-125
Number of pages18
JournalInformation Sciences
Volume333
DOIs
Publication statusPublished - 10 Mar 2016

Fingerprint

Chemotherapy
Control Design
Cancer
Drugs
Oncology
Toxicity
Model structures
Cells
Dose
Derivatives
Schedule
Cell
Biological systems
Evolutionary algorithms
Directly proportional
Differential equations
Control System
Scheduling
Availability
Mathematical models

Keywords

  • Cancer chemotherapy
  • Frequency response data
  • Killing cancerous cells
  • Measurement-based cancer control
  • PID controller design

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Theoretical Computer Science
  • Software
  • Computer Science Applications
  • Information Systems and Management
  • Artificial Intelligence

Cite this

A measurement-based control design approach for efficient cancer chemotherapy. / Khadraoui, Sofiane; Harrou, Fouzi; Nounou, Hazem; Nounou, Mohamed; Datta, Aniruddha; Bhattacharyya, Shankar P.

In: Information Sciences, Vol. 333, 10.03.2016, p. 108-125.

Research output: Contribution to journalArticle

Khadraoui, Sofiane ; Harrou, Fouzi ; Nounou, Hazem ; Nounou, Mohamed ; Datta, Aniruddha ; Bhattacharyya, Shankar P. / A measurement-based control design approach for efficient cancer chemotherapy. In: Information Sciences. 2016 ; Vol. 333. pp. 108-125.
@article{05d5eecc47734401a5a09ddf2be54e00,
title = "A measurement-based control design approach for efficient cancer chemotherapy",
abstract = "This paper presents a new measurement-based control design method for cancer chemotherapy. Cancer chemotherapy aims basically at simultaneously eradicating or significantly reducing the number of cancer cells and maintaining tolerable levels of drug concentration and toxicity. To achieve such aim, drugs are often injected into the patient's body according to a drug schedule specifying the drug dose and delivery time. Several strategies for planning cancer chemotherapy have been developed in the literature, where evolutionary algorithms have been applied to find optimal drug schedules of cancer treatment under constraints on some key treatment parameters such as drug concentration and toxic side effects. In such methods, the amount of drug doses, delivered in the body at each time during the treatment, does not depend on the current drug concentration, toxicity level, and/or number of cancer cells. Successful design of chemotherapy drug scheduling requires the availability of an accurate mathematical model that perfectly predicts the number of cancerous cells and describes effects of treatment. Several models with either complex or simple structures are available in the literature. Complex-structure models are proposed to deeply understand interactions between cancer and normal cells that affect the performance of the cancer chemotherapy. Nevertheless, such complex models are based on a high-order set of differential equations which can be difficult to solve. Simple-structure models, which are often obtained on the basis of some simplifying assumptions, can be viewed only as an approximation of the cancer system. Hence, designing chemotherapy drug schedules on the basis of simplified models may result in unsuccessful cancer treatment. Unlike conventional control strategies for cancer chemotherapy, our attempt in this paper is to address the problem of designing a control system for cancer treatment using a set of frequency-domain data. Hence, a two-degree-of-freedom PID (proportional-integral-derivative) control scheme is proposed to control cancer growth. These PID controllers are designed to simultaneously provide the optimal amount of drug doses to be delivered into the patient's body according to the current drug concentration and toxicity level, and maintain the drug concentration and toxicity levels within their pre-specified ranges. The proposed cancer control technique is validated through a first simulation example. Another example to control biological systems is also presented to show the feasibility of the proposed method. Simulation results obtained have demonstrated the capability of the proposed control scheme to address cancer chemotherapy problems.",
keywords = "Cancer chemotherapy, Frequency response data, Killing cancerous cells, Measurement-based cancer control, PID controller design",
author = "Sofiane Khadraoui and Fouzi Harrou and Hazem Nounou and Mohamed Nounou and Aniruddha Datta and Bhattacharyya, {Shankar P.}",
year = "2016",
month = "3",
day = "10",
doi = "10.1016/j.ins.2015.11.026",
language = "English",
volume = "333",
pages = "108--125",
journal = "Information Sciences",
issn = "0020-0255",
publisher = "Elsevier Inc.",

}

TY - JOUR

T1 - A measurement-based control design approach for efficient cancer chemotherapy

AU - Khadraoui, Sofiane

AU - Harrou, Fouzi

AU - Nounou, Hazem

AU - Nounou, Mohamed

AU - Datta, Aniruddha

AU - Bhattacharyya, Shankar P.

PY - 2016/3/10

Y1 - 2016/3/10

N2 - This paper presents a new measurement-based control design method for cancer chemotherapy. Cancer chemotherapy aims basically at simultaneously eradicating or significantly reducing the number of cancer cells and maintaining tolerable levels of drug concentration and toxicity. To achieve such aim, drugs are often injected into the patient's body according to a drug schedule specifying the drug dose and delivery time. Several strategies for planning cancer chemotherapy have been developed in the literature, where evolutionary algorithms have been applied to find optimal drug schedules of cancer treatment under constraints on some key treatment parameters such as drug concentration and toxic side effects. In such methods, the amount of drug doses, delivered in the body at each time during the treatment, does not depend on the current drug concentration, toxicity level, and/or number of cancer cells. Successful design of chemotherapy drug scheduling requires the availability of an accurate mathematical model that perfectly predicts the number of cancerous cells and describes effects of treatment. Several models with either complex or simple structures are available in the literature. Complex-structure models are proposed to deeply understand interactions between cancer and normal cells that affect the performance of the cancer chemotherapy. Nevertheless, such complex models are based on a high-order set of differential equations which can be difficult to solve. Simple-structure models, which are often obtained on the basis of some simplifying assumptions, can be viewed only as an approximation of the cancer system. Hence, designing chemotherapy drug schedules on the basis of simplified models may result in unsuccessful cancer treatment. Unlike conventional control strategies for cancer chemotherapy, our attempt in this paper is to address the problem of designing a control system for cancer treatment using a set of frequency-domain data. Hence, a two-degree-of-freedom PID (proportional-integral-derivative) control scheme is proposed to control cancer growth. These PID controllers are designed to simultaneously provide the optimal amount of drug doses to be delivered into the patient's body according to the current drug concentration and toxicity level, and maintain the drug concentration and toxicity levels within their pre-specified ranges. The proposed cancer control technique is validated through a first simulation example. Another example to control biological systems is also presented to show the feasibility of the proposed method. Simulation results obtained have demonstrated the capability of the proposed control scheme to address cancer chemotherapy problems.

AB - This paper presents a new measurement-based control design method for cancer chemotherapy. Cancer chemotherapy aims basically at simultaneously eradicating or significantly reducing the number of cancer cells and maintaining tolerable levels of drug concentration and toxicity. To achieve such aim, drugs are often injected into the patient's body according to a drug schedule specifying the drug dose and delivery time. Several strategies for planning cancer chemotherapy have been developed in the literature, where evolutionary algorithms have been applied to find optimal drug schedules of cancer treatment under constraints on some key treatment parameters such as drug concentration and toxic side effects. In such methods, the amount of drug doses, delivered in the body at each time during the treatment, does not depend on the current drug concentration, toxicity level, and/or number of cancer cells. Successful design of chemotherapy drug scheduling requires the availability of an accurate mathematical model that perfectly predicts the number of cancerous cells and describes effects of treatment. Several models with either complex or simple structures are available in the literature. Complex-structure models are proposed to deeply understand interactions between cancer and normal cells that affect the performance of the cancer chemotherapy. Nevertheless, such complex models are based on a high-order set of differential equations which can be difficult to solve. Simple-structure models, which are often obtained on the basis of some simplifying assumptions, can be viewed only as an approximation of the cancer system. Hence, designing chemotherapy drug schedules on the basis of simplified models may result in unsuccessful cancer treatment. Unlike conventional control strategies for cancer chemotherapy, our attempt in this paper is to address the problem of designing a control system for cancer treatment using a set of frequency-domain data. Hence, a two-degree-of-freedom PID (proportional-integral-derivative) control scheme is proposed to control cancer growth. These PID controllers are designed to simultaneously provide the optimal amount of drug doses to be delivered into the patient's body according to the current drug concentration and toxicity level, and maintain the drug concentration and toxicity levels within their pre-specified ranges. The proposed cancer control technique is validated through a first simulation example. Another example to control biological systems is also presented to show the feasibility of the proposed method. Simulation results obtained have demonstrated the capability of the proposed control scheme to address cancer chemotherapy problems.

KW - Cancer chemotherapy

KW - Frequency response data

KW - Killing cancerous cells

KW - Measurement-based cancer control

KW - PID controller design

UR - http://www.scopus.com/inward/record.url?scp=84952333634&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84952333634&partnerID=8YFLogxK

U2 - 10.1016/j.ins.2015.11.026

DO - 10.1016/j.ins.2015.11.026

M3 - Article

AN - SCOPUS:84952333634

VL - 333

SP - 108

EP - 125

JO - Information Sciences

JF - Information Sciences

SN - 0020-0255

ER -