Abstracts

Optimizing Responsive Neurostimulation Targeting Based on Interictal High-Frequency Oscillations and Phase-Amplitude Coupling

Abstract number : 1.271
Submission category : 3. Neurophysiology / 3G. Computational Analysis & Modeling of EEG
Year : 2025
Submission ID : 893
Source : www.aesnet.org
Presentation date : 12/6/2025 12:00:00 AM
Published date :

Authors :
Presenting Author: Sotaro Kanai, MD, PhD – Division of Pediatric Neurology, Department of Pediatrics, David Geffen School of Medicine at the University of California, Los Angeles, California, USA

Atsuro Daida, MD, PhD – Saitama Children's Medical Center, Saitama, Saitama, Japan
Yipeng Zhang, MS, PhD – Department of Electrical and Computer Engineering, University of California Los Angeles
Yuanyi Ding, MS – Department of Electrical and Computer Engineering, University of California Los Angeles
Tonmoy Monsoor, MS, PhD – Department of Electrical and Computer Engineering, University of California Los Angeles
Joe X. Qiao, MD, MS – Department of Radiology, UCLA Medical Center, David Geffen School of Medicine
Noriko Salamon, MD, PhD – Department of Radiology, UCLA Medical Center, David Geffen School of Medicine
Eric Ronne, MD – Division of Pediatric Neurology, Department of Pediatrics, UCLA Mattel Children's Hospital, David Geffen School of Medicine
Samuel S. Ahn, MD – UCLA
Shaun A. Hussain, MD, MS – Division of Pediatric Neurology, Department of Pediatrics, UCLA Mattel Children's Hospital, David Geffen School of Medicine
Raman Sankar, MD, PhD – Division of Pediatric Neurology, Department of Pediatrics, UCLA Mattel Children's Hospital, David Geffen School of Medicine
Aria Fallah, MD, MSc, MBA – Department of Neurosurgery, UCLA Medical Center, David Geffen School of Medicine
William Speier, PhD – Department of Radiological Sciences and Bioengineering, University of California Los Angeles
Vwani Roychowdhury, PhD – Department of Electrical and Computer Engineering, University of California Los Angeles
Jerome Engel Jr., MD, PhD – Department of Neurology, UCLA Medical Center, David Geffen School of Medicine
Richard J. Staba, PhD – Department of Neurology, UCLA Medical Center, David Geffen School of Medicine
Hiroki Nariai, MD, PhD, MS – Department of Pediatrics, Division of Pediatric Neurology, David Geffen School of Medicine at the University of California, Los Angeles, California, USA

Rationale: Responsive neurostimulation (RNS) offers an alternative treatment for patients with drug-resistant epilepsy (DRE) who are not candidates for resective surgery. However, determining optimal electrode placement remains a major clinical challenge, especially in pediatric-onset DRE, where seizure onset zones (SOZs) are often multifocal or poorly localized. High-frequency oscillations (HFOs) and delta-HFO phase-amplitude coupling (PAC), measured via the modulation index (MI), are well-established interictal biomarkers of epileptogenicity in surgical planning. Yet, their utility in guiding RNS implantation has not been validated. This study investigated whether RNS electrode targeting regions of increased interictal HFO and PAC results in better seizure outcomes compared to conventional SOZ-based targeting.

Methods: We retrospectively analyzed 18 patients (ages 6–28 years) with pediatric-onset DRE who underwent intracranial EEG (iEEG) and subsequent RNS implantation. Interictal HFOs were detected using both Montreal Neurological Institute (MNI) and short-time energy (STE) algorithms; delta-HFO PAC was quantified using MI. Biomarkers were processed using open-source tools (PyHFO and PACTv0.31) from slow-wave sleep recordings. For each patient, we calculated the weighted median distance between each iEEG channel and the nearest RNS electrode, using the proportion of HFOs or MI as weights. Patients were categorized as good (≥50% seizure reduction) or poor responders. Predictive performance was assessed using distance thresholding with leave-one-out cross-validation (LOOCV). Thalamic recordings and stimulation sites were analyzed separately. The study protocol was approved by the UCLA institutional review board.

Results: Good responders had significantly shorter weighted median distances for both HFOs and MI compared to poor responders (p < 0.0001, Figure 1A-C). Patient-wise analysis revealed significantly shorter distance only for MNI-HFO (p = 0.036, Figure 1D-F). RNS electrode placement within 20–30 mm of the peak HFO or PAC distribution yielded excellent predictive accuracy (AUC > 0.70), outperforming SOZ-based targeting (Figure 2). Among patients with thalamic RNS electrodes, those whose stimulation sites aligned with thalamic regions of elevated HFO/MI also showed favorable outcomes. MNI-based HFO detection showed stronger individual-level predictive performance than STE-based detection.

Conclusions: This study introduces the first interictal biomarker-based strategy for optimizing RNS targeting in DRE. Targeting regions with dense HFOs or delta-HFO PAC appears to improve outcomes more reliably than conventional SOZ-based placement. Because our method utilizes interictal data and fully open-source software, it is broadly applicable and immediately translatable to clinical practice. These findings support the feasibility of biomarker-guided RNS implantation and warrant prospective validation in larger, multi-institutional cohorts.

Funding: This study is supported by the Uehara Memorial Foundation for research abroad.

Neurophysiology