生物传热方程中灌注率函数的数值反演算法
2022-03-31曹庆发王泽文
曹庆发,胡 彬,万 殊,王泽文
生物传热方程中灌注率函数的数值反演算法
曹庆发,胡 彬,万 殊,*王泽文
(东华理工大学理学院,江西,南昌 330013)
本文研究了一类生物传热方程的灌注率函数反演问题。基于附加的非局部条件和有限差分的Crank-Nicolson方法,构造了重建灌注率函数的迭代算法;经进一步简化后,得到了反演灌注率的一个显格式。为克服计算的不稳定性,引入移动平均滤波方法对误差数据进行去噪,算例结果表明结合移动平均滤波去噪的数值反演算法是可行的,能有效反演出灌注率函数。
生物传热方程,灌注率,反问题,有限差分,移动平均
0 引言
生物医学传热研究不仅在理论上很重要,而且具有重要的实际应用价值[1]。例如,生物传热方程已被应用于模拟高热、血栓形成和血管硬化等研究中[2]。科学家提出了若干不同生物组织内的传热模型,例如最常用的是由Pennes提出的生物传热模型[3],以及文献[4-6]提出连续型生物传热模型。
由Pennes提出的生物传热方程[3]为
它反映了血液灌注率,故称其为血液灌注率函数。
和附加的非局部条件
不同于文献[11-12],本文受文献[16]中研究的启发,提出一种有限差分的反演算法,该方法无需事先将反问题转化为源项反演。本文接下来安排如下:第二小节基于Crank-Nicolson格式给出两种有限差分的数值反演算法;第三小节给出反问题的数值算例。
2 反问题的有限差分解法
2.1 反问题的数值解法
利用有限差分的Crank-Nicolson方法,将方程(2)离散为
由边界条件(3)-(5),有
对于附加条件(6),利用数值积分的复化梯形公式得
将(13)-(14)改写成矩阵形式为:
Step 3. 计算
上述差分格式的矩阵形式为
2.2 噪声处理的移动平均滤波
3 数值算例
算例1 考虑生物传热反问题:
其中精确解为
图1 算例1的反演结果对比
算例2 考虑生物传热反问题:
其中精确解为
图2 算例2的反演结果对比
数值算例的结果图1,图2表明所给出的算法是可行的,且迭代算法的数值反演效果更佳,特别是在第二个数值算例中算法体现了较强的抗噪能力,这可能是移动平均滤波对算例2的数据去噪效果更佳的缘故。该方法也可以推广到求解高维生物传热方程的相关反问题。
[1] Chato J C. Fundamentals of Bioheat Transfer[M]. Berlin: Springer-Verlag, 1990.
[2] Liu J, Xu L X. Boundary information based diagnostics on the thermal states of biological bodies[J]. International Journal of Heat and Mass Transfer, 2000,43(16): 2827-2839.
[3] Pennes H H. Analysis of tissue and arterial blood temperatures in the resting human forearm[J]. Journal of applied physiology, 1948, 1(2): 93-122.
[4] Wulff W. The energy conservation equation for living tissue[J]. IEEE transactions on biomedical engineering, 1974 (6): 494-495.
[5] Klinger H G. Heat transfer in perfused biological tissue—I: General theory[J]. Bulletin of Mathematical Biology, 1974, 36: 403-415.
[6] Chen M M, Holmes K R. Microvascular contributions in tissue heat transfer[J]. Annals of the New York Academy of Sciences, 1980, 335(1): 137-150.
[7] Grabski J K, Lesnic D, Johansson B T. Identification of a time-dependent bio-heat blood perfusion coefficient[J]. International Communications in Heat and Mass Transfer, 2016, 75: 218-222.
[8] Lin Y. An inverse problem for a class of quasilinear parabolic equations[J]. SIAM journal on mathematical analysis, 1991, 22(1): 146-156.
[9] Cannon J R, Lin Y, Wang S. Determination of a control parameter in a parabolic partial differential equation[J]. Journal of the Australian Mathematical Society, Series B Applied mathematics, 1991, 33: 149-163.
[10] Kerimov N B, Ismailov M I. An inverse coefficient problem for the heat equation in the case of nonlocal boundary conditions[J]. Journal of Mathematical Analysis & Applications, 2012, 396(2): 546-554.
[11] Wang S, Lin Y. A finite-difference solution to an inverse problem for determining a control function in a parabolic partial differential equation[J]. Inverse Problems, 1989, 5(4): 631-640.
[12] Ismailov M I, Kanca F. An inverse coefficient problem for a parabolic equation in the case of nonlocal boundary and overdetermination conditions[J]. Mathematical Methods in the Applied Sciences, 2011, 34(6): 692-702.
[13] Ismailov M I, Bazán F S V, Bedin L. Time-dependent perfusion coefficient estimation in a bioheat transfer problem[J]. Computer Physics Communications, 2018, 230: 50-58.
[14] Hazanee A, Lesnic D. Determination of a time-dependent coefficient in the bioheat equation[J].International Journal of Mechanical Sciences, 2014, 88: 259-266.
[15] Trucu D, Ingham D, Lesnic D. An inverse coefficient identification problem for the bio-heat equation[J]. Inverse Problems in Science and Engineering, 2009, 17(1): 65-83.
[16] Wang Z, Ruan Z, Huang H, et al. Determination of an unknown time-dependent heat source from A nonlocal measurement by finite difference method[J]. Acta Mathematicae Applicatae Sinica, English Series, 2020, 36(1): 151-165.
[17] 黄何露,王泽文,阮周生,等. 一类扩散方程寻源反问题的有限差分法[J].赣南师范大学学报,2018,39(3):20-23.
[18] 邱淑芳,王泽文,曾祥龙,等. 一类时间分数阶扩散方程中的源项反演解法[J]. 江西师范大学学报:自然科学版, 2018, 42(6): 610-615.
NUMERICAL METHOD FOR RECOVERING PERFUSION COEFFICIENT IN A BIOLOGICAL HEAT TRANSFER EQUATION
CAO Qing-fa, HU Bin, WAN Shu,*WANG Ze-wen
(School of Science, East China University of Technology, Nanchang, Jiangxi 330013, China)
The inversion of the perfusion coefficient function of a class of bioheat transfer equations is studied in this paper. Based on the additional non-local conditions and the Crank-Nicolson method of finite difference, an iterative algorithm for reconstructing the perfusion coefficient function is constructed; after further simplification, an explicit scheme for retrieving perfusion coefficient is obtained. In order to overcome the instability of calculation, the moving average filtering method is introduced to denoise the error data. The results of calculation examples show that numerical inversion algorithms combined with the moving average filtering denoising are feasible and effective for retrieving perfusion coefficient function.
bioheat transfer equation; perfusion coefficient; inverse problem; finite difference, moving average
1674-8085(2022)02-0022-06
O29
A
10.3969/j.issn.1674-8085.2022.02.004
2021-08-01;
2021-09-18
国家自然科学基金项目(11961002,11761007);江西省教育厅科技计划项目(GJJ170444);东华理工大学大学生科技创新基金项目
曹庆发(1996-),男,江西赣州人,硕士生,主要从事一般反问题的计算方法研究(E-mail:cqingfa58@163.com);
*王泽文(1974-),男,江西上饶人,教授,博士,主要从事一般反问题的计算方法研究(E-mail:zwwang6@163.com).