EEG University

Clarity AI Study Manuscript: Monitoring the Burden of Seizures in Critical Care with a Novel Machine Learning Method

Baharan Kamousi, Ceribell; Suganya Karunakaran, Ceribell; Kapil Gururangan, The Mount Sinai Hospital; et. al. | Accepted 6 October 2020

Ceribell with Clarity provides 24/7 EEG monitoring with consistently high performance for seizure burden assessment and bedside alert

• Of the 179 EEG recordings in which Claritγ detected no seizures, [Clarity had a] negative predictive value of 99%.
• Individual expert raters displayed variability in both sensitivity (range 20–89%) and specificity (range 94–99%).
• The sensitivity for identifying status epilepticus was 100%, [and] the specificity of the ≥ 90% seizure burden notification was 93.0%.