Jong Woo Lee

Background and ObjectivesThe EEG plays a crucial role in prognosticating outcomes for comatose patients after cardiac arrest (CA), but the precise probability of poor outcome (PPO) associated with ...

Objective: Sudden unexpected death in epilepsy (SUDEP) is the leading cause of death in otherwise healthy patients with epilepsy; about 70% occurs during sleep, and nearly 90% of patients are found prone (face‐down). Following a nocturnal seizure, terminal apnea occurs between 2.5 and 10.8 min postictally (median ~ 5 min), defining a narrow but actionable window for intervention. Although body‐position sensors are well established in sleep medicine and have been applied in epilepsy, no current product combines prone‐position detection with autonomous repositioning into a recovery (sideways) position. Methods: Korus is a smart mattress made of an array of inflatable cells. Korus consists of a sensor system to rapidly detect body change with high accuracy and a grid of expandable cells that reposition a patient from the prone to the recovery position. Results: Ten normative control subjects were recruited, nine of whom were systematically tested. Each Korus cell generated a lift of 454 kg/cell at 5 cm/s. The sensor array detected body position changes within 5 s. Manually controlling the cells to reposition the subject from prone to recovery was successful in 100% of attempts in an average of 21.75 (±5.85) seconds. Body position detection was 96.8% accurate in detecting the prone position and 92.4% accurate overall in detecting the correct body position (supine, prone, left, right). False body position detection rate was 2.3 events/h. There were no incorrect position detections when subjects switched into the prone position. Significance: This study demonstrates the feasibility of developing a smart mattress to detect the prone position and rapidly reposition subjects into a recovery position. Future development includes a control system to automate repositioning based on sensor data. The Korus device, when paired with a seizure detection device, may help lower the risk of night‐time SUDEP, though the magnitude of any benefit remains to be established. Plain Language Summary: Sudden unexpected death in epilepsy (SUDEP) most frequently occurs at night, and the prone position is a major risk factor. A device to autonomously reposition a patient to the recovery position may markedly reduce the rates of SUDEP. Here we describe the Korus, a smart mattress with the ability to reposition patients out of the prone position in their sleep.

Summary: Postanoxic encephalopathy is a common consequence of cardiac arrest, characterized by varying degrees of global cerebral hypoxic-ischemic injury. Despite advances in resuscitation science and critical care, neurologic outcome is essentially unchanged over the past decades. Although various treatment approaches to mitigate brain injury have been proposed, none of these has been unequivocally associated with improved neurologic outcome. This review provides an overview of the evidence from previous studies and ongoing clinical trials. Based on successes and limitations in recent research, we recommend EEG-based patient stratification and sufficiently powered subgroups for future trials. We also recommend outcome measures that provide more granularity than the current ordinal outcome scales, such as screening instruments for cognitive and emotional functioning. Adaptive randomized trial designs using a single master protocol and Bayesian statistics can provide an efficient platform for testing multiple treatments against a common control group, simultaneously. Treatments based on active stimulation of neuro- and synaptogenesis instead of prevention of secondary injury are promising.

Objective: Detection of seizures and rhythmic or periodic patterns (SRPPs) on electroencephalography (EEG) is crucial for the diagnosis and management of patients with neurological critical illness. Although automated detection methods are advancing, they require large, high‐quality, expert‐annotated datasets for training. However, expert annotation is limited by the availability of trained neurophysiologists. Crowdsourcing, or soliciting contributions from a large group of people, may present a potential solution. This study evaluates the feasibility of crowdsourcing annotations of short epochs of EEG recordings by comparing the performance of experts and non‐experts in identifying six SRPPs. Methods: We conducted an EEG scoring contest using a mobile app, involving expert and non‐expert participants. Non‐experts in our studies include physicians‐MD, medical students‐MS, nurse‐practitioner‐NP/physician‐assistant‐PA/pharmacists, NP‐students/PA‐students/pharmacy‐students/other‐healthcare‐students, and others. Performance was assessed using pairwise agreement and Fleiss' kappa between experts, and accuracy comparisons between experts and the crowd using individual and weighted majority votes. Results: A total of 1542 participants (8 experts and 1534 non‐experts) answered 478 834 questions across six SRPPs: seizures, generalized and lateralized periodic discharges (GPDs and LPDs), and generalized and lateralized rhythmic delta activity (GRDA LRDA), and “Other.” Using individual, non‐weighted votes, the crowd's performance was inferior to experts for overall and across six SRPP identification. Using weighted majority votes, the crowd was non‐inferior to experts for overall SRPP identification with accuracy of .70, 95% confidence interval [CI]: .69–.70 compared to expert's accuracy of .68, 95% CI: .68–.70. The crowd performed comparably or better than experts in identifying most SRPPs, except for LPDs and “Other”. No individual expert outperformed the crowd on overall metrics. Significance: This proof‐of‐concept highlights the promise of crowd reviewers for obtaining expert‐level annotations of SRPPs, which could potentially accelerate the development of large, diverse datasets for training automated detection algorithms. Challenges, such as varying calibration/test splits across crowd participants in the study and the absence of gold standard labels in the real‐world settings, remain to be addressed.

Summary: Rapid electroencephalography (EEG) devices are portable, easy-to-use systems that provide automated EEG interpretation to guide clinical decisions at the bedside. The scope of this review outlines the principles of automatic EEG analysis methods and the impact of recent developments in artificial intelligence and machine learning on these techniques in the context of seizure detection. The first generation of rapid EEG devices, primarily developed for anesthesia management, were tested for seizure detection; although raw EEG traces and power density arrays from these devices were helpful in seizure monitoring, their proprietary algorithms were not reliable for this purpose. The next generation of rapid EEG devices were principally aimed at seizure detection. In several large retrospective research studies, deep learning algorithms have demonstrated equal or superior proficiency as compared with human experts. Recent guidelines from the US Food and Drug Administration describe several guiding principles for this class of algorithms to be incorporated into medical devices. However, the “black box” nature of proprietary algorithms remains a concern because it hinders understanding of the device's limitations and potential pitfalls. We advocate for transparent reporting of essential software features to foster a constructive interaction between clinical users and rapid EEG device developers. This transparency is crucial for the continuous improvement of these devices, their acceptance among users, and ultimately, better patient care.

Rationale: Data on weight changes associated with Cenobamate (CNB) in patients with epilepsy is scarce. We performed a case‐control study because of concern for clinically significant weight loss.Methods: All patients exposed to CNB between its May in February 2020 until October 2023 with at least 180 days of exposure were evaluated. Baseline weight was assessed as the closest within 90 days prior to initiation of CNB. Trough and peak weights during CNB exposure and concomitant antiseizure medications (ASMs) were obtained.Results: A total of 332 patients obtained at least one dose of CNB during the study period, of whom 118 with accurate weight measurements were included in the analysis. A total of 33 patients (27.8%) were recorded as having experienced at least a 10% weight loss during CNB exposure; 16 patients (13.6%) had 10%–14.9% weight loss, 11 patients (9.3%) had 15%–19.9%, and 6 patients (5.1%) had ≥20%. A total of 9 (7.6%) patients had at least 10% weight gain and 1 patient had >20% weight gain. Median dose was 200 mg/d (IQR 50–400 mg/d); 27 of 33 patients had doses of ≥200 mg/d and experienced greater weight loss. There was a trend for male sex and duration to be associated with >10% weight loss. In four patients who underwent dose reduction or discontinuation due to weight loss, weight stabilized or increased without seizure exacerbation. Patients with greater initial weight were more likely to move into a lower BMI category (p < .01).Conclusions: A larger than expected number of patients in two centers had significant unintentional weight loss associated with CNB use. In a subset, the effect appeared to be reversible with minor dose reduction, even with severe weight loss. Systematic weight assessment should be considered in all patients on CNB, especially with telehealth. Clinicians should rule out contributing etiologies and consider CNB dose reduction if excessive weight loss occurs.