A fuzzy adaptive proportional-integral-derivative (PID) control strategy based on insulin-on-board control method of closed-loop insulin infusion

doi:10.3969/j.issn.2095-4344.1420

Abstract

Abstract:

BACKGROUND: Intelligent blood glucose control algorithm is an important part of closed loop artificial pancreas.

OBJECTIVE: To effectively control the blood glucose level of patients with type 1 diabetes mellitus, reduce the incidence of high or low blood glucose level, and propose a closed-loop insulin control algorithm based on fuzzy adaptive proportional-integral-derivative control strategy.

METHODS: According to the real-time monitoring data of continuous human blood glucose, proportional-integral-derivative algorithm was used to simulate the physiological transmission process of insulin secretion from human β cells. Then the parameters of the proportional-integral-derivative controller were continuously optimized. The amount of insulin-on-board was used to correct the insulin dose calculated by fuzzy adaptive proportional-integral-derivative algorithm to realize the optimal control of insulin pump.

RESULTS AND CONCLUSION: The performance of the proposed control algorithm was tested using the UVa/Padova stimulator jointly developed by the University of Virginia (USA) and University of Padova (Italy). Results showed that the insulin control algorithm significantly reduced hypoglycemia events, controlled blood glucose levels in the designated target range, increased the accuracy and effectiveness of insulin injection therapy, and greatly reduced the occurrence of various complications.

Key words: type 1 diabetes, artificial pancreas, fuzzy adaptive PID, blood glucose level, insulin, insulin-on-board, continuous glucose monitoring, UVa/Padova stimulator

CLC Number:

Yu Liling, Zhang Gangping, Liu Wenping, Xu Binfeng, Jin Haoyu. A fuzzy adaptive proportional-integral-derivative (PID) control strategy based on insulin-on-board control method of closed-loop insulin infusion[J]. Chinese Journal of Tissue Engineering Research, 2019, 23(30): 4816-4821.

Figures/Tables 6

目前，美国弗吉尼亚大学与意大利帕多氏联合开发的UVa/Padova仿真平台[16-17]，该仿真平台的仿真人体模型已被美国食品和药物管理局批准用于替代动物实验。该平台提供了虚拟的CGMS系统和胰岛素泵，用户只需要将自己的控制算法导入就可对任何虚拟患者进行血糖控制仿真实验。实验的血糖控制算法在该平台的学术版本进行了仿真测试，该学术版本包含30个虚拟的糖尿病患者数据(10例成年患者、10例青少年患者和10例儿童患者)，利用MATLAB软件的Simulink对实验中的算法进行仿真控制设计，将设计好的控制算法植入糖尿病模拟治疗测试软件，对算法进行性能测试。在仿真平台测试时，测试生成3 d的模拟场景，成人和青少年也使用同样的进餐方案。在大多数情况下，该文算法能够使得糖尿病患者的血糖浓度控制在目标范围(3.8-10 mmol/L)内具有高百分比。输入和输出变量的论域取值以下：e(k)、ec(k) ={-3，-2，-1，0，1，2，3}；?KP ={-0.5，-0.2，-0.1，0，0.1，0.2，0.5}，?KI ={-0.5，-0.2，-0.1，0，0.1, 0.2，0.5}，?KD ={-0.5，-0.2，-0.1，0，0.1，0.2，0.5}。该文控制器效果良好，10例成人患者的测试结果如图2所示，其中第一、二行代表没有进行血糖控制的效果，第三、四行代表该文方法的控制效果。从图上可看到10名受试者(成人01、02、03、04、05、06、07、08、09、10)血糖浓度在目标范围(3.8-10 mmol/L)内的百分比分别上升了42%，2%，41%，33%，68%，18%，89%，66%，22%，5%。图3显示了10例青少年的测试结果，从图上可看到10名受试者(青少年01、02、03、04、05、06、07、08、09、10)血糖浓度在目标范围内的百分比分别上升了51%，38%，56%，61%，62%，54%，45%，45%，65%，56%。对比成人的控制效果，青少年的控制稍微差些，这归因于青少年的血糖波动比成年人高得多。"

控制变异性网格分析用来评估所有患者的血糖浓度调节质量[18]，在评估闭环胰岛素控制算法及比较它们的性能方面发挥重要作用。假定对于每个受试者，在指定的时间段(例如1 d)内测量的糖尿病患者血糖浓度时间序列可用，控制变异性网格分析如下获得：对于每个受试者，绘制一个点，其X坐标代表最小血糖浓度，其Y坐标代表的所考虑时间段内的最大血糖浓度。网格内9个矩形区域定义如下：A区意味着精确控制，Lower B区意味着低血糖的良性偏差，Upper B区意味着高血糖的良性偏差，B区意味着良性控制偏差，Lower C意味着血糖浓度过度纠正，Upper C意味着低血糖过度纠正，Lower D意味着不能处理低血糖事件，Upper D意味着不能处理高血糖事件，E意味着错误的控制。图4显示了其中一组糖尿病患者的控制变异性网格分析，从图上可看到控制变异性网格分析中的所有点都位于A区和B区，意味着该文提出的血糖控制方法具有更好的血糖浓度调节控制。图5显示了10例成年人的仿真结果，SD代表标准偏差，用来描述各患者血糖值偏离平均血糖距离的平均数，它是离差平方和平均后的方根，实线代表虚拟成人糖尿病患者的平均血糖浓度，红色虚线代表血糖值在平均值上面波动变化。在虚拟成人糖尿病患者的测试中，该文提出的血糖浓度控制算法被证明了可改善糖尿病患者的血糖控制，同时能够降低患者低血糖发生率，平均98.5%的仿真时间内成年糖尿病患者的血糖浓度都保持在目标范围内。图6显示了10例青少年患者的平均血糖浓度和SD。该文方法能够跟踪患者血糖浓度的变化。在青少年糖尿病患者的测试过程中，血糖浓度在目标范围内占得比例平均为91.57%，并且在夜间血糖能够保持稳定。"

References

[1]Liu WP, Zhang GP, Yu LL, et al. Improved GPC Algorithm for Blood Glucose Control of Type 1 Diabetes. Artif Organs.2019;43(4):386-398.

[2]Hirsch IB.Clinical review: Realistic expectations and practical use of continuous glucose monitoring for the endocrinologist.J Clin Endocrinol Metab.2009;94(7):2232-2238.

[3]Cobelli C, Renard E, Kovatchev B. Artificial pancreas: past, present, future. Diabetes. 2011; 60: 2672-2682.

[4]Davey RJ, Low C,Jones TW,et al.Contribution of an intrinsic lag of continuous glucose monitoring systems to differences in measured and actual glucose concentrations changing at variable rates in vitro.J Diabetes Sci Technol.2010;4(6):1393-1399.

[5]Bagis A,Senberber H.ABC algorithm based PID controller design for higher order oscillatory systems.Elektron Elektrotech.2017;23(6):3-9.

[6]Soylu S,Danisman K.In silico testing of optimized Fuzzy P+D controller for artificial pancreas. Biocybern Biomed Eng.2018;38(2):399-408.

[7]Gondhalekar R,Dassau E,Doyle FJ 3rd.MPC design for rapid pump-attenuation and expedited hyperglycemia response to treat T1DM with an artificial pancreas.Proc Am Control Conf.2014;2014:4224-4230.

[8]Turksoy K,Quinn L,Littlejohn E,et al.Multivariable adaptive identification and control for artificial pancreas systems.IEEE Trans Biomed Eng. 2014;61:883-891.

[9]Turksoy K,Cinar A. Adaptive control of artificial pancreas systems-a review. J Healthc Eng. 2014; 5:1-22.

[10]Abbes IB,Richard PY,Lefebvre MA,et al.A closed-loop artificial pancreas using a proportional integral derivative with double phase lead controller based on a new nonlinear model of glucose metabolism.J Diabetes Sci Technol.2013;7:699-707.

[11]Steil GM,Rebrin K,Janowski R,et al.Modeling beta-cell insulin secretion–implications for closed-loop glucose homeostasis.Diabetes Technol Ther.2003;5(6):953-964.

[12]Eren M.Adaptive control strategy for regulation of blood glucose levels in patients with type 1 diabetes.J Proc Control.2009;19:1333-1346.

[13]Kovatchev B,Cobelli C,Renard E,et al.Multinational study of subcutaneous model-predictive closed-loop control in type 1 diabetes mellitus: summary of the results.J Diabetes Sci Technol. 2010;4(6): 1374-1381.

[14]Mauseth R,Hirsch IB, Bollyky J,et al.Use of a "fuzzy logic" controller in a closed-loop artificial pancreas.Diabetes Technol Ther. 2013;15(8): 628-633.

[15]Khooban MH, Abadi DNM, Alfi A, et al.Swarm optimization tuned Mamdani fuzzy controller for diabetes delayed model.Turk J Elec Eng Comp Sci.2013;21:2110-2126.

[16]Man CD, Micheletto F, Lv D,et al.The UVA/PADOVA Type 1 Diabetes Simulator: New Features.J Diabetes Sci Technol.2014;8(1):26-34.

[17]Visentin R,Man CD,Cobelli C.One-Day Bayesian Cloning of Type 1 Diabetes Subjects: Toward a Single-Day UVA/Padova Type 1 Diabetes Simulator.IEEE Trans Biomed Eng. 2016;63(11):2416-2424.

[18]闫朝峰.基于个体化参数的血糖控制策略设计[D].北京:北京化工大学,2014.

[19]Turksoy K,Samadi S,Feng J,et al.Meal-Detection in Patients with Type 1 Diabetes: A New Module for The Multivariable Adaptive Artificial Pancreas Control System.IEEE J Biomed Health Inform. 2016;20(1): 47-54.