Background and Objectives
The aim of this work was to test the accuracy of Persyst commercially available automated seizure detection in critical care EEG by comparing automated seizure detections to human review in a manually reviewed cohort and on a large scale.
Automated seizure detections (Persyst versions 12 and 13) were compared to human review in a pilot cohort of 229 seizures from 85 EEG records and then in an expanded cohort of 7,924 EEG records. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated for individual seizures (pilot cohort) and for entire records (pilot and expanded cohorts). We assessed EEG features associated with the accuracy of automated seizure detections.
In the pilot cohort, accuracy of automated detection for individual seizures was modest (sensitivity 0.50, PPV 0.60). At the record level (did the recording contain seizures or not?), sensitivity was higher (pilot cohort 0.78, expanded cohort 0.91), PPV was low (pilot cohort 0.40, expanded cohort 0.08), and NPV was high (pilot cohort 0.88, expanded cohort 0.97). Different software versions (version 12 vs 13) performed similarly. Sensitivity was higher for records containing focal-onset seizures compared to generalized-onset seizures (0.93 vs 0.85, p = 0.012).
In critical care continuous EEG recordings, automated detection of individual seizures had rates of both false negatives and false positives that bring into question its utility as a seizure alarm in clinical practice. At the level of entire EEG records, the absence of automated detections accurately predicted EEG records without true seizures. The true value of Persyst automated seizure detection appears to lie in triaging of low-risk EEGs.
Classification of Evidence
This study provides Class II evidence that an automated seizure detection program cannot accurately identify EEG records that contain seizures.