en
×

分享给微信好友或者朋友圈

使用微信“扫一扫”功能。
参考文献 1
Tennant F.The physiologic effects of pain on the endo-crine system.Pain and Therapy.2013,2(2):75~ 86
参考文献 2
Yanovski J A,Cutler G B J,Chrousos G P,et al.Corti-cotropin-releasing hormone stimulation following low-dose dexamethasone administration.A new test to distinguish Cushing′s syndrome from pseudo-Cushing′ s states.The Journal of the American Medical Association.1993,269(17):2232~ 2238
参考文献 3
J.Yeo K T,Babic N,Hannoush Z,Weiss R E.Endocrine testing protocols:hypo-thalamic pituitary adrenal axis.South Dartmouth:MDText.com,Inc.,2015
参考文献 4
Griep E N,Boersma J W,Lentjes E G,et al.Function ofthe hypothalamic-pituitary-adrenal axis inpatients with fi-bromyalgia and low back pain.The Journal of Rheumatolo-gy.1998,25(7):1374~ 1381
参考文献 5
Besedovsky H,Chrousos G,Del Rey A.The hypothala-mus pituitary adrenal axis(1st ed.).Amsterdam:Elsevier.2008.
参考文献 6
Generaal E,Vogelzangs N,Macfarlane G J,et al.Re-duced hypothalamic-pituitary-adrenal axis activity in chro-nic multi-site musculoskeletal pain:partly masked by de-pressive and anxiety disorders.BMC Musculoskeletal Dis-orders.2014,15(1):227
参考文献 7
Andersen M,Vinther F,Ottesen J T.Mathematical mod-eling of the Hypothalamic pituitary adrenal gland(HPA)axis,including hippocampal mechanisms.Mathematical Biosciences.2013,246(1):122~ 138
参考文献 8
Sriram K,Rodriguez-Fernandez M,Doyle F J III.Model-ing cortisol dynamics in the neuro-endocrine axis distin-guishes normal,depression,and post-traumatic stress dis-order(PTSD)in humans.PLoS Computational Biology.2012,8(2):e1002379
参考文献 9
Stephenson E R,Kojouharov H V.A mathematical model of skeletal muscle regeneration.Mathematical Methods in the Applied Sciences.2018,(41):18
参考文献 10
Xiao P,Duan L,Su J.The synaptic plasticity variability in a post-traumatic stress disorder model.In:Advances in cognitive neurodynamics.Singapore:Springerꎻ2016,603 ~ 610
参考文献 11
Prince K,Campbell J,Picton P,et al.A computational model of acute pain.International Journal of Simulation:Systems,Science and Technology.2005,6(9):1~ 10
参考文献 12
Savic D,Jelic S,Buric N.Stability of a general delay dif-ferential model of the hypothalamo-pituitary-adrenocortical system.International Journal of Bifurcation and Chaos.2006,16(10):3079~ 3085
参考文献 13
刘晶,曹秦禹,王子剑,等.耦合 Pre-Botzinger 复合体中神经元的反相簇放电模式及同步研究.动力学与控制学报,2017,15(1):75~ 79(Liu J,Cao Q Y,Wang Z J,et al.Study on anti-phase bursting mode and synchroniza-tion of coupled neurons in Pre-Bötzinger complex.Journal of Dynamics and Control,2017,15(1):75 ~ 79(in Chi-nese))
参考文献 14
刘赵凡,孙晓娟,李慧妍.噪声诱使模块化神经元网络产生随机多共振现象.动力学与控制学报,2019,17(2):191 ~ 196(Liu Z F,Sun X J,Li H Y.Stochastic multi-resonance induced by noise in a neuronal network of subnetworks.Journal of Dynamics and Control,2019,17(2):191~ 196(in Chinese))
参考文献 15
Douglas A J.Central noradrenergic mechanisms underlying acute stress responses of the Hypothalamo-pituitary-adre-nal axis:adaptations through pregnancy and lactation.Stress.2005,8(1):5~ 18
参考文献 16
Pruessner J C,Hellhammer D H,Kirschbaum C.Burn-out,perceived stress,and cortisol responses to awakening.Psychosomatic Medicine.1999,61(2):197~ 204
参考文献 17
Gatchel R J,Bevers K,Licciardone J C,et al.Transitio-ning from acute to chronic pain:an examination of different trajectories of low-back pain.Healthcare.2018,6(48):1 ~ 12
目录contents

    摘要

    在临床医学中,确定疼痛如何从急性转变为慢性是有效预防和治疗的关键.因此目前迫切需要一种定量和预测的方法来评估疼痛的分类,从而能够更好地干预治疗.本文采用计算机建模的方法,利用下丘脑-垂体-肾上腺轴(HPA)模型来模仿疼痛轨迹,我们根据 cortisol 的取值,把它分为高低两个状态,cortisol 取值高的状态用来模拟高强度的疼痛,取值低的状态用来模拟轻微疼痛或者没有痛苦.通过对其模拟数据进行研究,分析了急性慢性疼痛的转变.其仿真结果说明,神经计算科学在评估病人模型方面具有可行性和很大的潜力.

    Abstract

    Identifying how pain transitions from acute to chronic is critical in designing effective prevention and management techniques. So it is necessary to find a way for a quantitative and predictive method to evaluate how pain trajectories are classified and then how we can effectively intervene them. In this paper,we use HPA model to simulate the pain trajectories. The cortisol-level values were classified into two different groups,and they pres- ent high-state and low-state patterns. The higher-cortisol level was used to model the high-intensity pain,and the lower-level to represent the low-pain or no-pain. The results show the feasibility and the potential of developing such a broader model for patient evaluations.

    关键词

    疼痛HPA模型神经计算数值模拟

    Keywords

    painHPAmodelneuralcomputingsimulation

  • 0 引言

  • 在临床医学中,普遍认为疼痛时下丘脑垂体-肾上腺系统(HPA) 会发生异常,从而导致肾上腺皮质激素、皮质醇( cortisol)激素等的变化[1].文献 [2-6]总结对比了现有的测量HPA系统激素变化的不同方法,认为HPA轴在大脑平衡激素的过程中发挥着重要作用.因为HPA轴是一个重要的反映压力和疼痛指标的系统,所以很多测量数据,例如病人恢复轨迹,EEG数据,皮质醇变化水平等都可以通过HPA系统来研究数据的变化.这个过程可以用基于相关生物机制建立的神经计算模型来实现,从而能够更好地实时预测疼痛轨迹.

  • 事实上,在计算神经科学中,模型的数值计算结果已经被广泛应用于疾病的预测.例如基于HPA系统建立的数学模型,就被应用于抑郁症、创伤后应激障碍( PTSD) 以及疼痛等相关的研究中[7-10] .在文献[11]中,作者模拟了急性疼痛的发生机制, 演示了在模型中外界输入是如何从脊髓到达大脑并导致了疼痛,从而建立了生物学意义可信的数学模型.文献[12]通过对HPA模型的数值运算,解释了压力和疼痛的关系.因此,神经生物数学模型的数值模拟及对应的动力学分析对于研究复杂生物过程有积极的作用,是对临床和实验的补充[13,14].事实上,HPA轴的正常运转与否确实与疼痛和肌肉压力关系密切.在本文中,我们利用HPA数学模型来模拟不同程度的疼痛所对应的动力学机制,从而来预测将神经计算模型应用于疼痛诊断治疗的可行性,这是一个影响深远的医学力学交叉研究课题.

  • 1 模型选择

  • HPA系统在平衡大脑荷尔蒙水平,对抗压力方面起着重要的作用[12] .具体表现为:在对抗压力的过程中,随着时间的推移,含有神经内分泌的下丘脑室旁核会产生促肾上腺皮质激素释放激素(CRH),脑垂体前叶受CRH刺激分泌促肾上腺皮质激素(ACTH),而肾上腺皮质受到ACTH的刺激会产生皮质醇激素(cortisol).图1 为上述所描述的HPA工作原理.

  • 在此过程中,皮质醇激素作为一种与外界压力有关的主要激素,能影响身体的很多方面,尤其是对大脑的影响[5,15,16].基于如上生物背景, 文献 [17]建立了一套非线性微分方程组来模拟整个HPA系统,方程组如下所示

  • d[CRH]dt=kstecsKin2Kin2+[GR]n2-Vs3[CRH]Km1+[CRH]
    (1)
  • d[ACTH]dt=Kp2[CRH]Kin2Kin2+[GR]n2-VS4[ACTH]Km2+[ACTH]
    (2)
  • d[CORT]dt=KP3[ACTH]-VSS[CORT]Km3+[CORT]Kd3[CORT]
    (3)
  • VS2[GR]n1K1n1+[GR]n1
    (4)
  • 图1 下丘脑—垂体—肾上腺轴(HPA-axis)示意图

  • Fig.1 A Diagram of Hypothalamic Pituitary Adrenal(HPA)axis.

  • 此模型精确模拟了大脑的四个区域,通过激素的化学突触耦合,向其它区域传输兴奋性或抑制性的信号.在模型中,CRH表示促肾上腺皮质激素释放激素( CRH), ACTH表示促肾上腺皮质激素(ACTH),CORT表示皮质醇激素( coritsol).我们用MATLAB的ode45 方法来求解微分方程,几个关键参数取值见表1,其它具体参数意义及取值请参见文献[7].

  • 表1 HPA模型关键参数取值

  • Table 1 HPA model and parameters

  • 2 数值模拟与临床对应

  • 因为皮质醇激素( cortisol)在疼痛治疗中起关键作用,而在数学模型中,我们可以实时控制代表cortisol的参数变量,因此,我们用cortisol的时间历程图及动力学分析来模拟预测临床中的疼痛轨迹.

  • 在图2 中,我们计算了100 天内cortisol激素的变化曲线.在计算过程中,我们取了每个小时的数据,这样每天会包含24 个数据点.基于HPA系统内大脑激素随压力变化的特征,cortisol激素呈现随时间变化而周期震荡的模式,对应于在治疗过程中疼痛正在发生时的情况.在图2 中,cortisol的值维持在0.1344 m/DL到0.13465 m/DL之间.这些数据可以和疼痛治疗中的临床数据作比对,从而应用于验证和预测.因为病人可以实时监测皮质醇的分泌,从而可以进一步了解疼痛和皮质醇治疗的轨迹模式.

  • 图2100 天内皮质醇激素的数值模拟数据

  • Fig.2 Simulated data of cortisol level during a period of 100 days

  • 在图3 中,对应于图2 的cortisol变化轨迹,我们根据cortisol的取值,把它分为高低两个状态, cortisol取值高的状态用来模拟高强度的疼痛,取值低的状态用来模拟轻微疼痛或者没有痛苦.在这种情况下,我们把一天24 个小时的值取平均作为一个值,以天为单位,图3 给出了震荡模式下100 天内的一个cortisol值走势,我们以0.1345 m/DL为阈值分为高低两个状态.图3 这种状态可以用来模拟慢性和间歇性疼痛.

  • 在HPA模型中,反映突触连接性的参数在预测疼痛状态时是至关重要的.我们可以通过调节参数的变化,从而得到不同的动力学模式来对应不同的疼痛状态.在图4(a)中,我们将HPA模型中的参数Vs4的取值从15μg/d降到0.907μg/d,其它参数

  • 图3 高低两种状态的疼痛轨迹模拟

  • Fig.3 A simulated 2-state pain trajectory

  • 保持不变,即可得到这一状态.此种状态的cortisol时间历程图可以用来模拟急性疼痛的情形.从图中可以看到,在100 天的时间周期内,急性疼痛在1-2 周内迅速消退,从图4(a)中可以看到,皮质醇激素水平经过了前期两周的震荡后,很快恢复到一个平缓的取值水平,意味着急性疼痛有所好转.在图4(b)中,和图3 一样,我们依然把此过程中的corti-sol值归为三类,高低状态分别对应剧痛和痛苦减轻,平均值水平状态对应无痛(治愈)状态.取一天24 小时的平均值作为阈值,把100 天内的cortisol取值分为三个状态:当cortisol≥0.23986 m/DL时, 归为剧烈疼痛的日子;当cortisol≤0.23985 m/DL时,属于疼痛减轻的日子;而0.23986 m/DL>cortisol>0.23985 m/DL时,可视为疼痛已治愈.图4 模拟的这种状态是急性腰背痛发作的代表,即对应的疼痛轨迹会迅速收敛到治愈或无疼痛状态.

  • 在图5 中,HPA模型模拟了临床中所见的狂躁性疼痛(manic pain),在模型中对应参数Kstress取值从每天10.1μg/d升到30μg/d,其它参数不变.如图5 所示,在100 天的模拟中,一个高低状态的周期大概需要10 天左右.图5(a)计算得到的corti-sol时间历程轨迹对应实际中的反复疼痛模式,图5(b)为根据阈值归过类的疼痛轨迹.在此过程中,归类模式和图4 一样,根据cortisol值分为三个不同的组:剧烈疼痛,疼痛减轻和无疼痛(治愈)状态.首先计算一天24 小时的平均值,在此状态下介于0.23990 m/DL和0.23980 m/DL之间,然后以此为标准,分cortisol值为三类:当cortisol≥0.23990 m/DL时为剧烈疼痛日, 当0.23990 m/DL> cortisol> 0.23980 m/DL时为无痛或治愈时, 当cortisol ≤ 0.23980 m/DL时为痛苦减轻时.

  • 图4 急性疼痛所对应的cortisol轨迹模拟图

  • Fig.4 A simulated cortisol level data and pain trajectory for acute pain

  • 以上都为某个病人在一段时间内的疼痛轨迹模拟.接下来我们试图模拟一组病人的情形.与单个病人每人都有自己的激素变化水平不同的是,我们希望得到一组病人的cortisol值的平均分布状态.假设皮质醇激素的平均值是正规化的,这样所有的差异都是由于个体之间随机波动引起的.我们随机选取了20 组皮质醇激素递减的数据,见图6.

  • 图5 狂躁型疼痛所对应的Cortisol轨迹模拟图

  • Fig.5 A Simulated Cortisol Trajectory Data for a Manic-style Pain

  • 图6( a) 首先给出了全部的cortisol的时间历程;然后计算了这20 个病例的平均值,图6( b)计算了平均值对各个cortisol值的方差,在这个过程中;除了KStress是随机取值;其它参数都和图4 中的参数一样; Kstress的平均值为0.9922101μg/d; 方差值为0.2907101μg/d.然后我们照例把20 组数据的均值按照阈值(0.552 到0.556 之间)分为三

  • 图620 组病人的Cortisol

  • 值和所对应的疼痛轨迹Fig.6 The cortisol trajectory and pain trajectory modeling for a group of 20 patients

  • 类:当cortisol≥0.556 m/DL时为剧烈疼痛日,0.556 m/DL>cortisol>0.552 m/DL时为治愈状态,而corti-sol≤0.552 m/DL时为痛苦减轻时.图6( c) 模拟数据显示了对均值分类后疼痛病人从剧痛到恢复的一个过程.

  • 3 结论

  • 在本文中,HPA神经计算模型提供了对疼痛机制的深入了解.由计算分析可以看到,从急性疼痛到慢性疼痛的转变,以及疼痛轨迹的模式都可以用HPA模型很好地模拟.尽管目前由于模型仅依靠Cortisol激素进行模拟预测有局限性,但是如文中已有成果显示,这对阐明疼痛轨迹并进一步预测发展、干预治疗还是有重要意义的.接下来需要进一步做的是如何把这些结果尽快地应用到临床医学中去,为正在遭受痛苦的病人提供更好的服务.

  • 参考文献

    • [1] Tennant F.The physiologic effects of pain on the endo-crine system.Pain and Therapy.2013,2(2):75~ 86

    • [2] Yanovski J A,Cutler G B J,Chrousos G P,et al.Corti-cotropin-releasing hormone stimulation following low-dose dexamethasone administration.A new test to distinguish Cushing′s syndrome from pseudo-Cushing′ s states.The Journal of the American Medical Association.1993,269(17):2232~ 2238

    • [3] J.Yeo K T,Babic N,Hannoush Z,Weiss R E.Endocrine testing protocols:hypo-thalamic pituitary adrenal axis.South Dartmouth:MDText.com,Inc.,2015

    • [4] Griep E N,Boersma J W,Lentjes E G,et al.Function ofthe hypothalamic-pituitary-adrenal axis inpatients with fi-bromyalgia and low back pain.The Journal of Rheumatolo-gy.1998,25(7):1374~ 1381

    • [5] Besedovsky H,Chrousos G,Del Rey A.The hypothala-mus pituitary adrenal axis(1st ed.).Amsterdam:Elsevier.2008.

    • [6] Generaal E,Vogelzangs N,Macfarlane G J,et al.Re-duced hypothalamic-pituitary-adrenal axis activity in chro-nic multi-site musculoskeletal pain:partly masked by de-pressive and anxiety disorders.BMC Musculoskeletal Dis-orders.2014,15(1):227

    • [7] Andersen M,Vinther F,Ottesen J T.Mathematical mod-eling of the Hypothalamic pituitary adrenal gland(HPA)axis,including hippocampal mechanisms.Mathematical Biosciences.2013,246(1):122~ 138

    • [8] Sriram K,Rodriguez-Fernandez M,Doyle F J III.Model-ing cortisol dynamics in the neuro-endocrine axis distin-guishes normal,depression,and post-traumatic stress dis-order(PTSD)in humans.PLoS Computational Biology.2012,8(2):e1002379

    • [9] Stephenson E R,Kojouharov H V.A mathematical model of skeletal muscle regeneration.Mathematical Methods in the Applied Sciences.2018,(41):18

    • [10] Xiao P,Duan L,Su J.The synaptic plasticity variability in a post-traumatic stress disorder model.In:Advances in cognitive neurodynamics.Singapore:Springerꎻ2016,603 ~ 610

    • [11] Prince K,Campbell J,Picton P,et al.A computational model of acute pain.International Journal of Simulation:Systems,Science and Technology.2005,6(9):1~ 10

    • [12] Savic D,Jelic S,Buric N.Stability of a general delay dif-ferential model of the hypothalamo-pituitary-adrenocortical system.International Journal of Bifurcation and Chaos.2006,16(10):3079~ 3085

    • [13] 刘晶,曹秦禹,王子剑,等.耦合 Pre-Botzinger 复合体中神经元的反相簇放电模式及同步研究.动力学与控制学报,2017,15(1):75~ 79(Liu J,Cao Q Y,Wang Z J,et al.Study on anti-phase bursting mode and synchroniza-tion of coupled neurons in Pre-Bötzinger complex.Journal of Dynamics and Control,2017,15(1):75 ~ 79(in Chi-nese))

    • [14] 刘赵凡,孙晓娟,李慧妍.噪声诱使模块化神经元网络产生随机多共振现象.动力学与控制学报,2019,17(2):191 ~ 196(Liu Z F,Sun X J,Li H Y.Stochastic multi-resonance induced by noise in a neuronal network of subnetworks.Journal of Dynamics and Control,2019,17(2):191~ 196(in Chinese))

    • [15] Douglas A J.Central noradrenergic mechanisms underlying acute stress responses of the Hypothalamo-pituitary-adre-nal axis:adaptations through pregnancy and lactation.Stress.2005,8(1):5~ 18

    • [16] Pruessner J C,Hellhammer D H,Kirschbaum C.Burn-out,perceived stress,and cortisol responses to awakening.Psychosomatic Medicine.1999,61(2):197~ 204

    • [17] Gatchel R J,Bevers K,Licciardone J C,et al.Transitio-ning from acute to chronic pain:an examination of different trajectories of low-back pain.Healthcare.2018,6(48):1 ~ 12

  • 参考文献

    • [1] Tennant F.The physiologic effects of pain on the endo-crine system.Pain and Therapy.2013,2(2):75~ 86

    • [2] Yanovski J A,Cutler G B J,Chrousos G P,et al.Corti-cotropin-releasing hormone stimulation following low-dose dexamethasone administration.A new test to distinguish Cushing′s syndrome from pseudo-Cushing′ s states.The Journal of the American Medical Association.1993,269(17):2232~ 2238

    • [3] J.Yeo K T,Babic N,Hannoush Z,Weiss R E.Endocrine testing protocols:hypo-thalamic pituitary adrenal axis.South Dartmouth:MDText.com,Inc.,2015

    • [4] Griep E N,Boersma J W,Lentjes E G,et al.Function ofthe hypothalamic-pituitary-adrenal axis inpatients with fi-bromyalgia and low back pain.The Journal of Rheumatolo-gy.1998,25(7):1374~ 1381

    • [5] Besedovsky H,Chrousos G,Del Rey A.The hypothala-mus pituitary adrenal axis(1st ed.).Amsterdam:Elsevier.2008.

    • [6] Generaal E,Vogelzangs N,Macfarlane G J,et al.Re-duced hypothalamic-pituitary-adrenal axis activity in chro-nic multi-site musculoskeletal pain:partly masked by de-pressive and anxiety disorders.BMC Musculoskeletal Dis-orders.2014,15(1):227

    • [7] Andersen M,Vinther F,Ottesen J T.Mathematical mod-eling of the Hypothalamic pituitary adrenal gland(HPA)axis,including hippocampal mechanisms.Mathematical Biosciences.2013,246(1):122~ 138

    • [8] Sriram K,Rodriguez-Fernandez M,Doyle F J III.Model-ing cortisol dynamics in the neuro-endocrine axis distin-guishes normal,depression,and post-traumatic stress dis-order(PTSD)in humans.PLoS Computational Biology.2012,8(2):e1002379

    • [9] Stephenson E R,Kojouharov H V.A mathematical model of skeletal muscle regeneration.Mathematical Methods in the Applied Sciences.2018,(41):18

    • [10] Xiao P,Duan L,Su J.The synaptic plasticity variability in a post-traumatic stress disorder model.In:Advances in cognitive neurodynamics.Singapore:Springerꎻ2016,603 ~ 610

    • [11] Prince K,Campbell J,Picton P,et al.A computational model of acute pain.International Journal of Simulation:Systems,Science and Technology.2005,6(9):1~ 10

    • [12] Savic D,Jelic S,Buric N.Stability of a general delay dif-ferential model of the hypothalamo-pituitary-adrenocortical system.International Journal of Bifurcation and Chaos.2006,16(10):3079~ 3085

    • [13] 刘晶,曹秦禹,王子剑,等.耦合 Pre-Botzinger 复合体中神经元的反相簇放电模式及同步研究.动力学与控制学报,2017,15(1):75~ 79(Liu J,Cao Q Y,Wang Z J,et al.Study on anti-phase bursting mode and synchroniza-tion of coupled neurons in Pre-Bötzinger complex.Journal of Dynamics and Control,2017,15(1):75 ~ 79(in Chi-nese))

    • [14] 刘赵凡,孙晓娟,李慧妍.噪声诱使模块化神经元网络产生随机多共振现象.动力学与控制学报,2019,17(2):191 ~ 196(Liu Z F,Sun X J,Li H Y.Stochastic multi-resonance induced by noise in a neuronal network of subnetworks.Journal of Dynamics and Control,2019,17(2):191~ 196(in Chinese))

    • [15] Douglas A J.Central noradrenergic mechanisms underlying acute stress responses of the Hypothalamo-pituitary-adre-nal axis:adaptations through pregnancy and lactation.Stress.2005,8(1):5~ 18

    • [16] Pruessner J C,Hellhammer D H,Kirschbaum C.Burn-out,perceived stress,and cortisol responses to awakening.Psychosomatic Medicine.1999,61(2):197~ 204

    • [17] Gatchel R J,Bevers K,Licciardone J C,et al.Transitio-ning from acute to chronic pain:an examination of different trajectories of low-back pain.Healthcare.2018,6(48):1 ~ 12

  • 微信公众号二维码

    手机版网站二维码