Background and Objectives: Seizures and other seizure-like patterns of brain activity can harm the brain and contribute to in-hospital death, particularly when prolonged. However, experts qualified to interpret electroencephalography (EEG) data are scarce. Prior attempts to automate this task have been limited by small or inadequately labeled samples and have not convincingly demonstrated generalizable expert-level performance. There exists a critical unmet need for an automated method to classify seizures and other seizure-like events with expert-level reliability. This study was conducted to develop and validate a computer algorithm that matches the reliability and accuracy of experts in identifying seizures and seizure-like events, known as “ictal-interictal-injury-continuum” (IIIC) patterns on EEG, including seizures (SZ), lateralized and generalized periodic discharges (LPD, GPD), and lateralized and generalized rhythmic delta activity (LRDA, GRDA), and in differentiating these patterns from non-IIIC patterns. Methods: We used 6,095 scalp EEGs from 2,711 patients with and without IIIC events to train a deep neural network, SPaRCNet, to perform IIIC event classification. Independent training and test datasets were generated from 50,697 EEG segments, independently annotated by 20 fellowship-trained neurophysiologists. We assessed whether SPaRCNet performs at or above the sensitivity, specificity, precision, and calibration of fellowship-trained neurophysiologists for identifying IIIC events. Statistical performance was assessed via the calibration index, and by the percentage of experts whose operating points were below the model’s receiver operating characteristic curves (ROC) and precision recall curves (PRC) for the 6 pattern classes. Results: SPaRCNet matches or exceeds most experts in classifying IIIC events based on both calibration and discrimination metrics. For SZ, LPD, GPD, LRDA, GRDA, and “Other” classes, SPaRCNet exceeds the following percentages of 20 experts – ROC: 45%, 20%, 50%, 75%, 55%, 40%; PRC: 50%, 35%, 50%, 90%, 70%, 45%; and calibration: 95%, 100%, 95%, 100%, 100%, 80%, respectively. Discussion: SPaRCNet is the first algorithm to match expert performance in detecting seizures and other seizure-like events in a representative sample of EEGs. With further development, SPaRCNet may thus be a valuable tool for expedited review of EEGs.

We describe 2 previously healthy children who suffered disabling arterial ischemic strokes because of acute intracranial large vessel occlusion within 3 to 4 weeks of coronavirus disease 2019 (COVID-19) infection. Both children presented from communities with high COVID-19 case rates in the Southwest United States. An 8-year-old American Indian girl experienced severe iron deficiency anemia requiring blood transfusion and presented with bilateral middle cerebral artery (MCA) distribution strokes 3 weeks later. She underwent emergent mechanical thrombectomy of the left MCA with successful clot retrieval but experienced reocclusion of that artery 5 hours after intervention. She also had evidence of cerebral arteritis on catheter angiography and vessel wall imaging, and clot pathology revealed recently formed, unorganized platelet- and fibrin-rich thrombus with sparse clusters of erythrocytes, degenerated histiocytes, few eosinophils, and rare neutrophils. A 16-year old African American boy demonstrated evidence of arteritis on brain magnetic resonance angiography and serological markers of cardiac and renal injury accompanied by positive lupus anticoagulant antibodies. The children described in this report express clinical features inconsistent with focal cerebral arteriopathy, including elevated markers of systemic inflammation in both bilateral MCA strokes in one case and multiple organ system dysfunction in the other case. Neither patient fulfilled criteria for multisystem inflammatory syndrome in children, given absence of fever. These cases illustrate that systemic postinfectious arteritis with cerebrovascular involvement may complicate COVID-19 infection in previously healthy school-aged children, and their presentations may overlap but not fulfill criteria for multisystem inflammatory syndrome in children or focal cerebral arteriopathy.

Transcranial Doppler ultrasonography (TCD) is being used in many pediatric intensive care units (PICUs) to aid in the diagnosis and monitoring of children with known or suspected pathophysiological changes to cerebral hemodynamics. Standardized approaches to scanning protocols, interpretation, and documentation of TCD examinations in this setting are lacking. A panel of multidisciplinary clinicians with expertise in the use of TCD in the PICU undertook a three-round modified Delphi process to reach unanimous agreement on 34 statements and then create practice recommendations for TCD use in the PICU. Use of these recommendations will help to ensure that high quality TCD images are captured, interpreted, and reported using standard nomenclature. Furthermore, use will aid in ensuring reproducible and meaningful study results between TCD practitioners and across PICUs.

Background and Objectives: The validity of brain monitoring using electroencephalography (EEG), particularly to guide care in patients with acute or critical illness, requires that experts can reliably identify seizures and other potentially harmful rhythmic and periodic brain activity, collectively referred to as “ictal-interictal-injury continuum" (IIIC). Prior inter-rater reliability (IRR) studies are limited by small samples and selection bias. This study was conducted to assess the reliability of experts in identifying IIIC. Methods: This prospective analysis included 30 experts with subspecialty clinical neurophysiology training from 18 institutions. Experts independently scored varying numbers of ten-second EEG segments as: “Seizure (SZ)”, “Lateralized Periodic Discharges (LPD)”, “Generalized Periodic Discharges (GPD)”, “Lateralized Rhythmic Delta Activity (LRDA)”, “Generalized Rhythmic Delta Activity (GRDA)”, or “Other”. EEGs were performed for clinical indications at Massachusetts General Hospital between 2006 to 2020. Primary outcome measures were pairwise IRR (average percent agreement (PA) between pairs of experts) and majority IRR (average PA with group consensus) for each class; and beyond chance agreement (κ). Secondary outcomes were calibration of expert scoring to group consensus, and latent trait analysis to investigate contributions of bias and noise to scoring variability. Results: Among 2,711 EEGs, 49% were from females, and median (IQR) age was 55 (41). In total experts scored 50,697 EEG segments; the median [range] number scored by each expert was 6,287.5 [1,002, 45,267]. Overall pairwise IRR was moderate (PA 52%, κ 42%), and majority IRR was substantial (PA 65%, κ 61%). Noise-bias analysis demonstrated that a single underlying receiver operating curve can account for most variation in experts9 false positive vs true positive characteristics (median [range] of variance explained (R²): 95 [93, 98]%), and for most variation in experts’ precision vs sensitivity characteristics (R²: 75 [59, 89]%). Thus, variation between experts is mostly attributable not to differences in expertise, but rather to variation in decision thresholds. Discussion: Our results provide precise estimates of expert reliability from a large and diverse sample, and a parsimonious theory to explain the origin of disagreements between experts. The results also establish a standard for how well an automated IIIC classifier must perform to match experts. Classification of Evidence: This study provides Class II evidence that independent expert review reliably identifies ictal-interictal injury continuum patterns on EEG compared to expert consensus.