基于电导的杏仁核模型：焦虑与抑郁中威胁加工的机制研究
A Conductance Based Amygdala Model of Threat Processing in Anxiety and Depression
AI总结：
本研究构建了基于电导的杏仁核-下丘脑-心血管通路模型，关联心理调节因子与心血管反应，为焦虑抑郁的数字表型和临床解读提供了可解释的机械框架。
AI中文摘要：
焦虑和抑郁障碍日益被视为连续的应激调节维度上的功能失调，但现有计算方法很少将可解释的环路水平机制与自主生理联系起来。方法：本研究开发了一种机械框架，将杏仁核功能失调与心血管应激反应关联，用于数字表型和临床解读。我们构建了一个紧凑的、含9个方程的、基于电导的杏仁核-下丘脑-心血管通路模型，该框架扩展了Hodgkin-Huxley形式体系，加入了3个基于临床的调节剂：应对能力、感知应激负荷和前额叶调节强度。模型采用缓慢的、依赖历史的内部状态、自适应阈值、分级威胁识别以及压力反射耦合的下丘脑整合，来生成心率和血压轨迹。结果：单一闭环系统的稳定运行状态呈现出强、中、弱三种不同的调节机制。鲁棒分析显示，随机变异性和参数扰动均保持了机制间的分离；消融研究确定自适应阈值是驱动定量机制分离的主要机制。模拟的心血管反应均处于已报道的应激生理范围内，且在三个独立数据集上的外部评估显示其与现实世界应激相关的自主模式具有强一致性。结论：紧凑的机械模型可在单一可解释框架内将心理测量调节剂、杏仁核兴奋性与下游心血管输出关联起来。意义：本研究为基于机制的数字表型、患者特异性应激监测以及未来用于心理健康决策支持的数字孪生方法提供了计算可行的基础。
英文摘要：
Anxiety and depressive disorders are increasingly viewed as dysregulations along continuous stress-regulatory dimensions. However, existing computational approaches seldom connect interpretable circuit level mechanisms to autonomic physiology. Methods: This study develops a mechanistic framework that links amygdala dysregulation to cardiovascular stress responses for digital phenotyping and clinical interpretation. We formulated a compact, nine equation, conductance based model of the amygdala hypothalamus cardiovascular pathway. The framework extends Hodgkin Huxley formalism with three clinically grounded modulators: coping capacity, perceived stress load, and prefrontal regulatory strength. A slow, history-dependent internal state, adaptive thresholding, graded threat acknowledgement, and baroreflex coupled hypothalamic integration were used to generate heart rate and blood pressure trajectories. Results: Distinct strong, moderate, and weak regulatory regimes emerged as stable operating states of a single closed-loop system. Robust analyses showed that stochastic variability and parameter perturbation preserved regime separation, while ablation studies identified the adaptive threshold as the principal mechanism driving quantitative regime separation. Simulated cardiovascular responses remained within reported stress physiology ranges. Furthermore, external evaluation across three independent datasets supported robust agreement with real world, stress related autonomic patterns. Conclusion: A compact, mechanistic model can jointly link psychometric modulators, amygdala excitability, and downstream cardiovascular output within a single, interpretable framework. Significance: This work provides a computationally tractable basis for mechanism informed digital phenotyping, patient specific stress monitoring, and future digital twin approaches for mental health decision support.
