Unmet Need
Every year, around 5 million people are diagnosed with epilepsy globally. About 30% of the global epileptic population have medically refractory epilepsy (MRE) with seizures that cannot be controlled with medication. MRE patients have two treatment options: surgical resection of or electrical stimulation of the seizure onset zone (SOZ). Both treatments can be lifesaving but rely on accurate identification of the SOZ. With current techniques of identification, patients must undergo prolonged hospital stays for observation of seizures, and clinicians generally have to visually inspect, channel by channel, hours of intracranial electroencephalography (iEEG) data looking for abnormalities that indicate epileptogenicity. These techniques are subjective and time consuming, and still result in poor surgical outcomes, in which the success rates of these treatments are still only 50% because of inaccurate SOZ localization. Thus, there is a need for a method that can efficiently and accurately localize the SOZ to improve patient surgical outcomes.
Technology Overview
The inventors have created patient-specific dynamical network models (DNMs) that can be used to characterize the dynamics of the brain network activity and provide quantitative metrics to be used in precisely localizing the SOZ. The network models are built from the evoked responses of patients that have undergone evaluation with single-pulse electrical stimulation. The DNMs are linear, time-invariant (LTI) state-space and single input multioutput transfer function models. Using properties of DNMs such as reachability and system gain, the models can accurately replicate the neural evoked responses within each patient and identify the SOZ, early spread, and irritative zones with greater accuracy than the time consuming, subjective localization techniques. This product, in the future, will be developed into a software tool that aids clinicians in identifying the SOZ. The clinicians will be able to upload the patients iEEG data and the software can process and analyze the data through the creation of DNM models. This technology will give clinicians greater insight into the network dynamics of a patient’s epilepsy, enabling a more precise surgical resection and ultimately improve surgical outcomes.
Stage of Development
The models have been created and are undergoing further verification and validation testing.
Publications
Kamali, Golnoosh et al. “Localizing the seizure onset zone from single pulse electrical stimulation responses using transfer function models.” Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference vol. 2020 (2020): 2524-2527. doi:10.1109/EMBC44109.2020.9175954
Smith, Rachel J et al. “State-space models of evoked potentials to localize the seizure onset zone.” Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference vol. 2020 (2020): 2528-2531. doi:10.1109/EMBC44109.2020.9176697
Rachel J Smith, Mark A Hays, Golnoosh Kamali, Christopher Coogan, Nathan E Crone, Joon Y Kang, Sridevi V Sarma, Stimulating native seizures with neural resonance: a new approach to localize the seizure onset zone, Brain, Volume 145, Issue 11, November 2022, Pages 3886–3900, https://doi.org/10.1093/brain/awac214
Kamali G, June-Smith R, Hays M, Coogan C, Crone NE, Kang JY, Sarma SV. Transfer Function Models for the Localization of Seizure Onset Zone from Cortico-Cortical Evoked Potentials. Frontiers Neurology. 10 December 2020, PMID: 33362689, PMCID: PMC7758451, DOI: 10.3389/fneur.2020.579961
