Intuitionistic Fuzzy Control of Different Strategies for Cancer Treatment

El-Garawany, Ahmed Hamdy; El-Brawany, Mohamed A

Intuitionistic Fuzzy Control of Different Strategies for Cancer Treatment

Document Type : Original Article

Authors

¹ industrial electronics and control department , faculty of electronic engineering, menoufia university

² Department of Industrial Electronics, Faculty of electronic engineering, Menoufia. University

Abstract

Chemotherapy is one of the most effective treatment methods for cancer patients. An intuitionistic fuzzy logic control (IFLC) of chemotherapy drug delivery system is presented in this paper. An Ant Colony Optimization (ACO) algorithm is 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 our controller by considering maximum allowed level constraints on both drug infusion dose and drug-dose toxicity. The obtained results from the optimized control system are compared with the previous studies. The application of the developed controller resulted in lowering the number of remaining cancer cells to 0.1856. with the highest performance index of 29.31.

Highlights

An intuitionistic fuzzy controller has been utilized to improve the drug infusion control system using. cancer patient model. The objective of controller to kill maximum number of remain cell and minimizing the side-effects of the drug. This has been achieved by adding an optimizer (ACO) to tune the PID-IFLC controller parameters. In this way, the ACO optimization is built to boost the search area and avoid local optima points. Drug dose, toxicity and concentration of drugs were always below the human's allowable limits. The future study will include the use of the established controller in animal studies using a commercial chemotherapy drugs infusion system.

Keywords

Main Subjects

• Biomedical Systems Engineering and Applications

Full Text

Cancer is a fetal disease, where number of abnormal cells start to be out of control and unfold into encompassing vital tissues. Cancer is one of the world's leading causes of mortality rates and death, with approximately 1,762,450 new cancer cases and 606,880 cancer deaths are estimated to happen in the United States [1] in 2019 . There are many techniques current used to tackle cancer, like surgery, radiation and chemotherapy.

Chemotherapy is an effective treatment type for fast spreading cancer. On the other hand, both surgery and radiation therapy remove, kill, or damage cancer cells in a specific area [2]. Use of chemotherapy drugs interrupts the cancer cell growth. Nevertheless, chemotherapy has two remarkable drawbacks that are drug resistance and the toxicity of drug [3]. All chemotherapy not only kill the cancerous cells, but they also have a bad effect on normal cells and destroy them. In contrast, inappropriate chemotherapeutic drugs can cause mutation of tumor cells.

The injection of chemotherapeutic agents into vein of the patient is referred to the intravenous chemotherapy clinical treatment. While drug dose scheduling plays a significant role to achieve an equilibrium between sparing the normal cells and killing the cancer cells with minimum level of toxicity and drug resistance impact [4]. The traditional intravenous infusion systems consist of a fluid container, administrator set, and a clamp to control the flow rate from the set to the patient. The major difficulty with this infusion system is that the flow cannot be accurately controlled and are always prone for human errors. To overcome this inaccuracy problems, electronic drug delivery systems are often used to regulate infused drug dose during chemotherapy sessions [5].

Mathematical modelling and control methods can be used to identify appropriate drug dose schedules for cancer treatment. Swierniak et al [6] and Araujo et al [7] provide reviews of how mathematical models were used to develop chemotherapeutic treatment schedules.

Drug resistance and toxicity, as already mentioned, are two significant issues that limit chemotherapy treatments. However, some earlier studies only took one of them into account. Batmani and Khaloozadeh [8] found certain optimal treatment regimens for cancer patients. But in their study, they did not consider the issue of drug resistance. Khaloozadeh et al. [9] achieved the best drug regimens taking into account two target functions to minimize the number of tumor cells and to protect normal cells. However, the drug resistance was also ignored in their models.

Martin [10] introduced an optimal chemotherapy drug scheduling model and solved it by applying nonlinear programming techniques. But he found difficulty while using these nonlinear techniques and hard to get a simplistic optimal control solution for all cancer chemotherapy drug scheduling problems. Tan et al [11] proposed a distributed evolutionary software that obtains automated solutions to the complex problem for the scheduling of chemotherapies. Liang et al. [12] updated the response model for chemotherapy in [10], because the drug toxicity modelling did not conform to relevant clinical work. This research team has been successful in achieving better outcomes by optimizing the schedule of chemotherapeutic agents using developed evolutionary algorithm; namely adaptive elitist-population-based genetic algorithms (AEGA) under specific toxicity tolerance [13]. Batmani et al. [14] extended the model proposed by Westman et al [15] to forecast the dynamics of tumor growth in the presence of chemo treatment and then used Non-dominated genetic sorting to fix the

problem of bi-objective optimization. Other methods have

also been used for the treatment of cancer chemotherapy,

such as improved immune algorithms [16] and the Multi-

Objective Evolutionary Approach (MOEA) [17].

But these open-loop systems for medication delivery can't

cope with patient variability. Moreover, they can't handle

any unforeseen health circumstance, such as a rise in

drug toxicity in the human body. In contrast, closed loop

systems [18-20] contain a feedback mechanism, more

accurate, robust in the existence of non-linearity and

external noise sources. According to these features, closed

loop control systems present a good alternative to safely

achieve cancer chemotherapy treatment goals. Algoul et

al. [21] applied PID and IPD controllers with a multiobjective

genetic algorithm in the closed loop

chemotherapy system to adjust controller parameters. The

optimized PID controller was applied successfully to

Martin’s chemotherapy response model. However, one

suitable solution for PID parameters must be carefully

chosen from several optimal Pareto solution sets.

Khadraoui et al [22] applied two PID controllers on a

modified Martin’s chemotherapy model to control both

toxicity and drug concentration in patient body. But the

authors in [21, 22] used conventional controller that

usually does not obtain satisfactory performance under

parameter variations [23].

Fuzzy logic controllers can work with imprecise inputs,

handles non-linearities and is more robust than

conventional controllers[24]. Fuzzy PID controllers have

self-tuning ability and on-line adaptation to nonlinear,

time varying, and uncertain systems [25]. However, fuzzy

sets suffer from the restriction of the uncertain element

because the lack of information. Uncertainty is an

inseparable aspect of medical problems [26]. Among

extensions of fuzzy sets, Atanassov’s intuitionistic fuzzy

sets [27] provide an intuitive structure for dealing with

ambiguity from imprecise knowledge by taking nonmembership

into account as well as membership values.

In this study, a PID-like IFLC controller is utilized to

control chemotherapy drug delivery system in a closed

loop. An ACO optimization is used to tune the

controller’s parameters to achieve the best results from

drug infusion process. Also, the control system is

designed to achieve both clinical and cancer therapy

needs.

The following is the organization of this paper. Section 2

explains the cancer chemotherapy drug scheduling model

and the controller design in the closed loop infusion

system. Section 3 presents the results of the developed

control system and compares them with the results of

previous works. Discussion is given in Section 4. Section

5 presents conclusion and comments on future work.

References

6. REFERENCES
[1] R. L. Siegel, K. D. Miller, and A. Jemal, "Cancer statistics, 2019," CA: a cancer journal for clinicians, vol. 69, no. 1, pp. 7-34, 2019.
[2] K. A. Pearlstein, R. Basak, and R. C. Chen, "Comparative Effectiveness of Prostate Cancer Treatment Options: Limitations of Retrospective Analysis of Cancer Registry Data," International Journal of Radiation Oncology* Biology* Physics, vol. 103, no. 5, pp. 1053-1057, 2019.
[3] U. Ledzewicz, S. Wang, H. Schattler, N. Andre, M. A. Heng, and E. Pasquier, "On drug resistance and metronomic chemotherapy: A mathematical modeling and optimal control approach," (in eng), Math Biosci Eng, vol. 14, no. 1, pp. 217-235, Feb 01 2017
[4] K. El-Husseiny, H. Motawei, and M. S. Ali, "Continuous Low-Dose Oral Cyclophosphamide and Methotrexate as Maintenance Therapy in Patients With Advanced Ovarian Carcinoma After Complete Clinical Response to Platinum and Paclitaxel Chemotherapy," (in eng), Int J Gynecol Cancer, vol. 26, no. 3, pp. 437-42, Mar 2016.
[5] R. L. Hill et al., "Comparison of drug delivery with autoinjector versus manual prefilled syringe and between three different autoinjector devices administered in pig thigh," (in eng), Med Devices (Auckl), vol. 9, pp. 257-66, 2016.
[6] A. Swierniak, M. Kimmel, and J. Smieja, "Mathematical modeling as a tool for planning anticancer therapy," European Journal of Pharmacology, vol. 625, pp. 108–121, 2009.
[7] R. P. Araujo and D. L. McElwain, "A history of the study of solid tumour growth: the contribution of mathematical modelling," (in eng), Bull Math Biol, Review vol. 66, no. 5, pp. 1039-91, Sep 2004.
[8] Y. Batmani and H. Khaloozadeh, "Optimal chemotherapy in cancer treatment: state-dependent Riccati equation control and extended Kalman filter, Optimal Control Appl," Optimal Control Applications and Methods, vol. 34, no. 5, pp. 562–577, 2013.
[9] H. Khaloozadeh, P. Yazdanbakhsh, and F. Homaei-shandiz, "The optimal dose of CAF regimen in adjuvant chemotherapy for breast cancer patients at stage IIB," (in eng), Math Biosci, vol. 213, no. 2, pp. 151-8, Jun 2008.
[10] R. B. Martin, "Optimal-Control Drug Scheduling of Cancer-Chemotherapy," (in English), Automatica, vol. 28, no. 6, pp. 1113-1123, Nov 1992.
[11] K. C. Tan, E. F. Khor, J. Cai, C. M. Heng, and T. H. Lee, "Automating the drug scheduling of cancer chemotherapy via evolutionary computation," (in English), Artif Intell Med, vol. 25, no. 2, pp. 169-185, Jun 2002.
[12] Y. Liang, K. S. Leung, and T. S. Mok, "A novel evolutionary drug scheduling model in cancer chemotherapy," (in eng), IEEE Trans Inf Technol Biomed, Research Support, Non-U.S. Gov't vol. 10, no. 2, pp. 237-45, Apr 2006.
[13] Y. Liang, K. S. Leung, and T. S. K. Mok, "Evolutionary drug scheduling models with different toxicity metabolism in cancer chemotherapy," (in English), Appl Soft Comput, vol. 8, no. 1, pp. 140-149, Jan 2008.
[14] Y. Batmani and H. Khaloozadeh, "Optimal drug regimens in cancer chemotherapy: a multi-objective approach," (in eng), Comput Biol Med, vol. 43, no. 12, pp. 2089-95, Dec 2013.
[15] Westman, B. R. Fabijonas, D. L. Kern, and F. B. Hanson, "Probabilistic Rate Compartment Cancer Model: Alternate Versus Traditional Chemotherapy Scheduling," Lecture Notes in Control and Information Sciences, Springer-Verlag, New York, vol. 2002, pp. 491–506, 2002.
[16] J.-T. Tsai, W.-H. Ho, and Y.-M. Chen, "Optimized Drug Scheduling for Cancer Chemotherapy using Improved Immune Algorithm," International Journal of Innovative Computing, Information and Control, vol. 9, pp. 2821-2838, 2013.
[17] Y. Batmani and H. Khaloozadeh, "Optimal drug regimens in cancer chemotherapy: A multi-objective approach," Computers in Biology and Medicine, vol. 43, no. 12, pp. 2089-2095, 2013.
[18] Y. Liu et al., "Closed-Loop Control Better than Open-Loop Control of Profofol TCI Guided by BIS: A Randomized, Controlled, Multicenter Clinical Trial to Evaluate the CONCERT-CL Closed-Loop System,"

(in English), Plos One, vol. 10, no. 4, Apr 17 2015.
[19] K. Kamalesh, P. S. P. Reddy, and J. S. Vinodhini, "Open loop and closed loop comparision for single phase Cyclo converter (PID controller)," in 2016 Second International Conference on Science Technology Engineering and Management (ICONSTEM), pp. 481-484,, 30-31 March 2016.
[20] K. H. Ang, G. Chong, and Y. Li, "PID control system analysis, design, and technology," (in English), Ieee T Contr Syst T, vol. 13, no. 4, pp. 559-576, Jul 2005.
[21] S. Algoul, M. S. Alam, M. A. Hossain, and M. A. Majumder, "Multi-objective optimal chemotherapy control model for cancer treatment," (in eng), Med Biol Eng Comput, vol. 49, no. 1, pp. 51-65, Jan 2011.
[22] S. Khadraouia, F. Harroub, H. N. Nounouc, M. N. Nounoub, A. Dattad, and S. P. Bhattacharyyad, "A measurement-based control design approach for efficient cancer chemotherapy," Information Sciences, vol. 333, pp. 108-125, 2016.
[23] P. Sooraksa and G. R. Chen, "On comparison of a conventional proportional-integral plus derivative controller versus a fuzzy proportional-integral plus derivative controller: A case study of subsystem failure," (in English), Ieee Icit' 02: 2002 Ieee International Conference on Industrial Technology, Vols I and Ii, Proceedings, pp. 205-207, 2002.
[24] M.-M. Frăţilă, V. Vlădăreanu, V.-A. Grosu, M. S. Munteanu, and Y. Feng, "Autonomous Mobile Robot Navigation Based On Fuzzy Logic CONTROL," Acta Electrotehnica, vol. 60, 2019.
[25] M. M. Hammad, A. K. Elshenawy, and M. I. El Singaby, "Trajectory following and stabilization control of fully actuated AUV using inverse kinematics and self-tuning fuzzy PID," (in eng), Plos One, vol. 12, no. 7, p. e0179611, 2017.
[26] P. Ejegwa and B. Onasanya, "Improved intuitionistic fuzzy composite relation and its application to medical diagnostic process," Note IFS, vol. 25, no. 1, pp. 43-58, 2019.
[27] H. Garg and K. Kumar, "Linguistic interval-valued Atanassov intuitionistic fuzzy sets and their applications to group decision-making problems," IEEE Transactions on Fuzzy Systems, 2019.
[28] R. Luus, F. Hartig, and F. J. Keil, "Optimal Drug Scheduling of Cancer-Chemotherapy by Direct Search Optimization," (in English), Hung J Ind Chem, vol. 23, no. 1, pp. 55-58, 1995.
[29] S. Algoul, M. S. Alam, M. A. Hossain, and M. A. A. Majumder, "Multi-objective optimal chemotherapy control model for cancer treatment," Medical & Biological Engineering & Computing, journal article vol. 49, no. 1, pp. 51-65, 2011.
[30] Y. Liang, K. S. Leung, and T. S. K. Mok, "A novel evolutionary drug scheduling model in cancer chemotherapy," (in English), Ieee T Inf Technol B, vol. 10, no. 2, pp. 237-245, Apr 2006.
[31] L. Preziosi, "Mathematical models to aid in cancer research," (in Italian), B Unione Mat Ital, vol. 8B, no. 1, pp. 55-76, Feb 2005.
[32] M. Dean, T. Fojo, and S. Bates, "Tumour stem cells and drug resistance," (in eng), Nat Rev Cancer, Review vol. 5, no. 4, pp. 275-84, Apr 2005.
[33] X.-h. Yuan, H.-x. Li, and C. Zhang, "The theory of intuitionistic fuzzy sets based on the intuitionistic fuzzy special sets," Information Sciences, vol. 27, pp. 284-298, 2014/09/01/ 2014.
[34] A. Arya and S. P. Yadav, "Development of intuitionistic fuzzy super-efficiency slack based measure with an application to health sector," Computers & Industrial Engineering, vol. 115, pp. 368-380, 2018/01/01/ 2018.
[35] K. T. Atanassov, "On the Concept of Intuitionistic Fuzzy Sets," in On Intuitionistic Fuzzy Sets Theory. Berlin, Heidelberg: Springer Berlin Heidelberg, pp. 1-16, 2012.
[36] F. Camastra et al., "A fuzzy decision system for genetically modified plant environmental risk assessment using Mamdani inference," Expert Systems with Applications, vol. 42, no. 3, pp. 1710-1716, 2015/02/15/ 2015.
[37] M. Akram, S. Shahzad, A. Butt, and A. Khaliq, "Intuitionistic Fuzzy Logic Control for Heater Fans," Mathematics in Computer Science, journal article vol. 7, no. 3, pp. 367-378, September 01 2013.
[38] M. E. Karar, "A Simulation Study of Adaptive Force Controller for Medical Robotic Liver Ultrasound Guidance," Arabian Journal for Science and Engineering, journal article vol. 43, no. 8, pp. 4229-4238, August 01 2018.
[39] M. Dorigo and T. Stützle, "Ant colony optimization: overview and recent advances," in Handbook of metaheuristics: Springer, 2019.
[40] M. Guntsch, M. Middendorf, and H. Schmeck, "An ant colony optimization approach to dynamic TSP," in Proceedings of the 3rd Annual Conference on Genetic and Evolutionary Computation: Morgan Kaufmann Publishers Inc., pp. 860-867, 2001.
[41] D. Merkle and M. Middendorf, "Studies on the dynamics of ant colony optimization algorithms," in Proceedings of the 4th Annual Conference on Genetic and Evolutionary Computation, 2002: Morgan Kaufmann Publishers Inc., pp. 105-112.
[42] J. Uthayakumar, N. Metawa, K. Shankar, and S. Lakshmanaprabu, "Financial crisis prediction model using ant colony optimization," International Journal of Information Management, vol. 50, pp. 538-556, 2020.