用智能算法优化脑深部电刺激电极配置，提升精准度并应对不确定性。
In Silico Study for Optimizing Intensity and Focality Electrode Configurations for Directional DBS Under Uncertainty Using Metaheuristic L1L1 Method
- 采用L1L1智能优化方法，融合电极位置与组织特性不确定性。
- 在8和40接触电极上实现稀疏且精准的刺激模式，抑制非目标区域电流扩散。
- 适合神经外科医生用于个性化脑深部刺激方案设计。
随着深部脑刺激（DBS）向方向性电极和基于优化的电流导向发展，电极接触配置的选择变得复杂。本研究将配置选择建模为靶区激活与电极电流之间的逆映射问题，采用元启发式L1范数正则化L1拟合（L1L1）方法，该方法整合了电极放置、组织特性及前向建模假设带来的场域不确定性。框架引入场域扰动，并通过敏感性可行性准则限制可控制域，基于完整电极模型的有限元公式构建。对8接触和40接触电极的电流分布进行了优化，评估指标包括聚焦电流密度、干扰电流密度及其比值，在安全性和稀疏性约束下进行测试。结果表明，无论在无噪声还是有噪声场域中（噪声模拟组织激活体积内衰减），该方法均生成了跨扰动水平的稀疏、空间选择性刺激模式；超参数优化得到双极或多极配置。相比仅产生严格双极配置的互易原理（RP）和更分散解的Tikhonov正则最小二乘法（TLS），L1L1实现了稀疏与多极模式间的可控过渡，有效集中刺激于靶区，尤其在噪声条件下显著抑制非预期电流扩散。结论：L1L1能辅助专家优化DBS配置，通过直接将不确定性纳入优化过程，提供鲁棒且可解释的电流导向能力，适应前向模型变异。
原文摘要 · Abstract (English)
Background and Objective: As Deep Brain Stimulation (DBS) advances toward directional leads and optimization-based current steering, selecting electrode contact configurations becomes complex. This study formulates configuration selection as an inverse mapping between target activation and electrode currents using metaheuristic L1-norm regularized L1-norm fitting (L1L1). L1L1 incorporates lead-field uncertainty arising from electrode placement, tissue properties, and forward modeling assumptions. Methods: The framework introduces lead-field perturbations and restricts the controllable domain through a sensitivity-based feasibility criterion within a finite element formulation derived using the Complete Electrode Model. Current distributions were optimized for 8- and 40-contact leads. Performance was evaluated using focused current density, nuisance current density, and their ratio under safety and sparsity constraints. Results: L1L1 was evaluated using noiseless and noisy lead fields, with noise selected to reflect attenuation within the volume of tissue activated. The method produced sparse, spatially selective stimulation patterns across perturbation levels. Hyperparameter optimization yielded bipolar or multipolar configurations. Compared with the Reciprocity Principle (RP), which produced strictly bipolar configurations, and Tikhonov-regularized least squares (TLS), which produced more distributed solutions, L1L1 enabled controlled transitions between sparse and multipolar patterns. It concentrated stimulation within the target while limiting unintended current spread, particularly under noisy conditions. Conclusions: L1L1 can assist specialists in optimizing DBS configurations. By incorporating uncertainty directly into optimization, it provides robust and interpretable current steering across lead configurations while accounting for forward-model variability.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。