Literature-Guided Minimax Optimization of Virtual Epilepsy Neurostimulation
2026-06-03 • Machine Learning
Machine Learning
AI summaryⓘ
The authors developed a computer-based method to design better epilepsy treatments that work well even for patients who react differently to brain stimulation. They combined information from many scientific papers, brain activity simulations, and an advanced optimization process to find the best treatment settings that minimize the worst seizure outcomes in virtual patients. While their method improved some treatment parameters noticeably in simulations, improvements for clinical-style brain stimulation were smaller and less convincing across a group of virtual patients. The authors emphasize that their work is a computer simulation test rather than proof the approach works in real patients.
epilepsyneuromodulationcomputational modelThe Virtual BrainEpileptorblack-box optimizationworst-case optimizationvirtual patientsneurostimulationliterature mining
Authors
Cathy Liu
Abstract
Computational models of epilepsy promise patient-specific treatment design, but most optimization workflows still search for parameters that perform well on average. In neuromodulation, this is a weak target: a protocol that improves the mean response can still fail in the patient whose network is least tolerant to stimulation. We present a literature-guided minimax pipeline that couples PubMed-scale hypothesis extraction, The Virtual Brain (TVB) Epileptor simulations, and large-language-model-guided black-box optimization. The optimizer proposes either intrinsic model-control parameters or clinically interpretable external-stimulation protocols; TVB evaluates each proposal across sampled virtual patients; and the objective maximizes worst-case reward, defined as the negative variance of simulated seizure activity. In the intrinsic model-control experiment, the best archived parameter set improved worst-case reward from -0.5285 to -0.3182, a 39.8% gain over baseline. The clinical-style external-stimulation search produced a much smaller worst-case improvement (1.7%), and a 20-patient virtual cohort showed no aggregate benefit (p=0.9019), despite a 55% responder rate and a positive temporal-lobe subgroup signal. The study should be read as an in silico proof of concept for robust, literature-aware neurostimulation design, not as clinical evidence.