Authors :
Presenting Author: Mehmet Kadipasaoglu, MD, PhD – McGovern Medical School, UTHealth
Adeel Ilyas, MD – McGovern Medical School, University of Texas Health Science Center at Houstont
Nitin Tandon, MD – McGovern Medical School, University of Texas Health Science Center at Houston
Rationale:
Approximately one-third of patients with epilepsy have drug-refractory epilepsy (DRE) and are potential candidates for surgical intervention. Stereo-electroencephalography (SEEG) enables sampling of deep and eloquent cortical, which requires the spatial integration of complex, multimodal datasets in the context of each patient’s complex cortical anatomy. Current methods for generating and synthesizing these data demand extensive time and cognitive effort, requiring dedicated personnel to operate complex software pipelines. To address this unmet clinical need, we have developed a clinical decision-support system, EnCompass, that automates the patient-specific data processing pathway. The EnCompass pipeline includes: (1) automated generation of patient-specific anatomical cortical/subcortical models; (2) co-registration, localization, and labeling of SEEG electrodes relative to these models; and (3) dynamic visualization of SEEG-derived ictal onset activity mapped onto patient-specific anatomical surfaces. In this fashion, EnCompass facilitates a seamless method for rapid visualization 4-dimensional (4-D) SEEG-derived ictal activations on patient-specific cortical anatomy, with the aim of enhancing the efficiency and accuracy of surgical planning.
Methods:
We retrospectively reviewed five adult patients who underwent SEEG-guided surgery for eloquent neocortical epilepsy between January and December 2024 at a single tertiary epilepsy center, in whom EnCompass was employed as an intraoperative navigational adjunct. Onset regions visualized in Encompass were compared to conventional SEEG-defined onset electrodes and final surgical resection zones. Seizure outcomes in this cohort were reviewed at >6 months postoperatively.
Results:
Five patients (median age 22.5 years) with eloquent neocortical epilepsies (aphasic or sensorimotor seizures) underwent SEEG-guided resections with EnCompass assistance. All patients were seizure free (Engel Class I) at six months (median follow-up 10.7 months). In all cases, EnCompass-derived onset regions were more spatially circumscribed than conventional SEEG-onset analyses, while maintaining close concordance with the final resection zone. Notably, EnCompass provided additional intraoperative guidance in complex cases, including resections near speech and Rolandic cortex where early ictal spread complicated localization using conventional SEEG review alone.
Conclusions:
EnCompass provides an integrated, automated, and efficient platform for seamless visualization of patient-specific SEEG-derived ictal onset activity within a 4-D anatomical framework. In this initial retrospective series, EnCompass enhanced spatial targeting of resection zones relative to conventional SEEG interpretation and demonstrated potential to meaningfully inform surgical decision-making in anatomically complex cases. These findings support EnCompass’ utility as an adjunct to standard workflows, streamlining the patient evaluation process while reducing cognitive and time demands on providers. Further prospective studies are warranted to assess its broader impact on surgical efficiency, cognitive workload, and long-term clinical outcomes.
Funding:
NIH U01 NS098981; NIH R01 DC014589