Digital Biomarkers to Screen for Absence Epilepsy on Visually Normal EEG Segments
Abstract number :
1.301
Submission category :
3. Neurophysiology / 3G. Computational Analysis & Modeling of EEG
Year :
2024
Submission ID :
737
Source :
www.aesnet.org
Presentation date :
12/7/2024 12:00:00 AM
Published date :
Authors :
Presenting Author: Saeid Sadeghian, MD – Boston Children's Hospital, 300 Longwood Ave, Boston, MA 02115, USA
Michele Jackson, BA – Boston Childrens Hospital
Lillian Voke, BS – UMass Chan Medical School
Tobias Loddenkemper, MD – Boston Children's Hospital, 300 Longwood Ave, Boston, MA 02115, USA
William Bosl, PhD – Computational Health Informatics Program, Boston Children's Hospital, Harvard Medical School, Boston, MA, USA, Clinical Neuroinformatics & AI Laboratory, The Data Institute, University of San Francisco, San Francisco, CA, USA
Rationale: Non-motor seizures, including absence seizures, are challenging to identify, and diagnostic delays may occur, potentially delaying treatment and impacting quality of life and cognitive function.1 We developed a screening test to help diagnose absence seizures on visually normal electroencephalogram (EEG).
Methods: We evaluated patients with absence seizures (1 to 21 years old) and age-matched controls without seizures or epilepsy who underwent an EEG at Boston Children's Hospital from 2010 to 2024. A clinical neurophysiologist selected visually normal 30 s awake and sleep EEG segments. We deployed linear and nonlinear methods, including recurrence quantification analysis (RQA) and sample entropy, to differentiate between EEG segments of patients with and without absence seizures. We calculated frequency bands (d-, d+, q, a, b, g, g+) using data from 18 out of 19 scalps sensors according to the standard 10-20 arrangement; the Cz sensor was not included in the calculation, for use as future reference point to compute 13 dynamic measures. To reduce the large number of individual measures (7*18*13 à 1638), we arranged data into a tensor structure with axes corresponding to dynamical measures, scalp location, frequency band, and subjects in the population. Supervised tensor factorization extracted 3 latent factors that were input to a Random Forest classifier (scikit-learn Python package). 5-fold cross-validation was used for both tensor factorization and classification between absence and control patients.
Results: We analyzed 161 absence patients (66.4% female, mean age: 10.1 yrs) and 162 controls (53.7% female, mean age: 9.6 yrs). Absence patients consisted of Childhood Absence Epilepsy (CAE, 94), Juvenile Absence Epilepsy (JAE, 39), and patients with generalized epilepsy and non-syndromic absence seizures (28). Our classifier differed between absence seizure patients and patients without epilepsy (p< 6.6e-4), including the separation of CAE or JAE groups from controls (Table 1). Nonlinear features differentiated subgroups of absence patients with high specificity (Table 1). CP tensor factorization assessed the contribution of raw factors from the latent factors (Figure 1). Most prominent raw measures included Sample Entropy, largest Lyapunov exponent, Determinism, Laminarity, and maximum recurrence line length. Scalp locations are dominated by frontal sensors with contributions by O1, T7, and P7. Lower frequencies in the delta range contributed most.
Neurophysiology