Seizure Assessment and Forecasting With Efficient Rapid-EEG: A Retrospective Multicenter Comparative Effectiveness Study.

BACKGROUND AND OBJECTIVES: Approximately 30% of critically ill patients have seizures, and more than half of these seizures do not have an overt clinical correlate. EEG is needed to avoid missing seizures and prevent overtreatment with antiseizure medications. Conventional-EEG (cEEG) resources are logistically constrained and unable to meet their growing demand for seizure detection even in highly developed centers. Brief EEG screening with the validated 2HELPS2B algorithm was proposed as a method to triage cEEG resources, but it is hampered by cEEG requirements, primarily EEG technologists. Seizure risk-stratification using reduced time-to-application rapid response-EEG (rrEEG) systems (∼5 minutes) could be a solution. We assessed the noninferiority of the 2HELPS2B score on a 1-hour rrEEG compared to cEEG. METHODS: A multicenter retrospective EEG diagnostic accuracy study was conducted from October 1, 2021, to July 31, 2022. Chart and EEG review performed with consecutive sampling at 4 tertiary care centers, included records of patients ≥18 years old, from January 1, 2018, to June 20, 2022. Monte Carlo simulation power analysis yielded n = 500 rrEEG; for secondary outcomes n = 500 cEEG and propensity-score covariate matching was planned. Primary outcome, noninferiority of rrEEG for seizure risk prediction, was assessed per area under the receiver operator characteristic curve (AUC). Noninferiority margin (0.05) was based on the 2HELPS2B validation study. RESULTS: A total of 240 rrEEG with follow-on cEEG were obtained. Median age was 64 (interquartile range 22); 42% were female. 2HELPS2B on a 1-hour rrEEG met noninferiority to cEEG (AUC 0.85, 95% CI 0.78-0.90, p = 0.001). Secondary endpoints of comparison with a matched contemporaneous cEEG showed no significant difference in AUC (0.89, 95% CI 0.83-0.94, p = 0.31); in false negative rate for the 2HELPS2B = 0 group (p = 1.0) rrEEG (0.021, 95% CI 0-0.062), cEEG (0.016, 95% CI 0-0.048); nor in survival analyses. DISCUSSION: 2HELPS2B on 1-hour rrEEG is noninferior to cEEG for seizure prediction. Patients with low-risk (2HELPS2B = 0) may be able to forgo prolonged cEEG, allowing for increased monitoring of at-risk patients. CLASSIFICATION OF EVIDENCE: This study provides Class II evidence that rrEEG is noninferior to cEEG in calculating the 2HELPS2B score to predict seizure risk.

From Sleep Patterns to Heart Rhythms: Predicting Atrial Fibrillation from Overnight Polysomnograms.

Background: Atrial fibrillation (AF) is often asymptomatic and thus under-observed. Given the high risks of stroke and heart failure among patients with AF, early prediction and effective management are crucial. Importantly, obstructive sleep apnea is highly prevalent among AF patients (60-90%); therefore, electrocardiogram (ECG) analysis from polysomnography (PSG), a standard diagnostic tool for subjects with suspected sleep apnea, presents a unique opportunity for the early prediction of AF. Our goal is to identify individuals at a high risk of developing AF in the future from a single-lead ECG recorded during standard PSGs. Methods: We analyzed 18,782 single-lead ECG recordings from 13,609 subjects at Massachusetts General Hospital, identifying AF presence using ICD-9/10 codes in medical records. Our dataset comprises 15,913 recordings without a medical record for AF and 2,056 recordings from patients who were first diagnosed with AF between 1 day to 15 years after the PSG recording. The PSG data were partitioned into training, validation, and test cohorts. In the first phase, a signal quality index (SQI) was calculated in 30-second windows and those with SQI < 0.95 were removed. From each remaining window, 150 hand-crafted features were extracted from time, frequency, time-frequency domains, and phase-space reconstructions of the ECG. A compilation of 12 statistical features summarized these window-specific features per recording, resulting in 1,800 features. We then updated a pre-trained deep neural network and data from the PhysioNet Challenge 2021 using transfer-learning to discriminate between recordings with and without AF using the same Challenge data. The model was applied to the PSG ECGs in 16-second windows to generate the probability of AF for each window. From the resultant probability sequence, 13 statistical features were extracted. Subsequently, we trained a shallow neural network to predict future AF using the extracted ECG and probability features. Results: On the test set, our model demonstrated a sensitivity of 0.67, specificity of 0.81, and precision of 0.3 for predicting AF. Further, survival analysis for AF outcomes, using the log-rank test, revealed a hazard ratio of 8.36 (p-value of 1.93 × 10 -52 ). Conclusions: Our proposed ECG analysis method, utilizing overnight PSG data, shows promise in AF prediction despite a modest precision indicating the presence of false positive cases. This approach could potentially enable low-cost screening and proactive treatment for high-risk patients. Ongoing refinement, such as integrating additional physiological parameters could significantly reduce false positives, enhancing its clinical utility and accuracy.

The predictive validity of a Brain Care Score for dementia and stroke: data from the UK Biobank cohort.

Introduction: The 21-point Brain Care Score (BCS) was developed through a modified Delphi process in partnership with practitioners and patients to promote behavior changes and lifestyle choices in order to sustainably reduce the risk of dementia and stroke. We aimed to assess the associations of the BCS with risk of incident dementia and stroke. Methods: The BCS was derived from the United Kingdom Biobank (UKB) baseline evaluation for participants aged 40-69 years, recruited between 2006-2010. Associations of BCS and risk of subsequent incident dementia and stroke were estimated using Cox proportional hazard regressions, adjusted for sex assigned at birth and stratified by age groups at baseline. Results: The BCS (median: 12; IQR:11-14) was derived for 398,990 UKB participants (mean age: 57; females: 54%). There were 5,354 incident cases of dementia and 7,259 incident cases of stroke recorded during a median follow-up of 12.5 years. A five-point higher BCS at baseline was associated with a 59% (95%CI: 40-72%) lower risk of dementia among participants aged <50. Among those aged 50-59, the figure was 32% (95%CI: 20-42%) and 8% (95%CI: 2-14%) for those aged >59 years. A five-point higher BCS was associated with a 48% (95%CI: 39-56%) lower risk of stroke among participants aged <50, 52% (95%CI, 47-56%) among those aged 50-59, and 33% (95%CI, 29-37%) among those aged >59. Discussion: The BCS has clinically relevant and statistically significant associations with risk of dementia and stroke in approximately 0.4 million UK people. Future research includes investigating the feasibility, adaptability and implementation of the BCS for patients and providers worldwide.

Self-Directed Home-Based Dim-Light Melatonin Onset Collection: The Circadia Pilot Study.

Study Objectives: To test the feasibility of a novel at-home salivary Dim Light Melatonin Onset (DLMO) assessment protocol to measure the endogenous circadian phase of 10 individuals ( 1 Advanced Sleep-Wake Phase Disorder patient (ASWPD), 4 Delayed Sleep-Wake Phase Disorder patients (DSWPD), and 5 controls). Methods: The study involved 10 participants (sex at birth: females = 9; male= 1), who ranged between 27 to 63 years old, with an average age of 38 years old. Our study population consisted of 7 individuals who identified as white and 3 who identified as Asian. Our participants were diverse in gender identity (woman = 7, male = 1, transgender = 1, nonbinary = 1, none = 1).The study tracked the sleep and activity patterns of 10 individuals over a 5-6 week period using self-reported online sleep diaries and objective actigraphy data. Participants completed two self-directed DLMO assessments, approximately one week apart, adhering to objective compliance measures. Participants completed the study entirely remotely: they completed all sleep diaries and other evaluations online and were mailed a kit with all materials needed to perform the actigraphy and at-home sample collections. Results: Salivary DLMO times were calculated for 8/10 participants using the Hockeystick method. DLMO times were on average 3 hours and 18 minutes earlier than self-reported sleep onset times (DSPD: 12:04 AM, controls: 9:55 PM.) Among the 6 participants for whom we calculated two separate DLMO times, DLMOs 1 and 2 were 96% correlated (p<0.0005.). Conclusions: Our results indicate that self-directed, at-home DLMO assessments are feasible and accurate. The current protocol may serve as a framework to reliably assess circadian phase in both clinical and general populations.

Improving automated diagnosis of epilepsy from EEGs beyond IEDs.

Objective.Clinical diagnosis of epilepsy relies partially on identifying interictal epileptiform discharges (IEDs) in scalp electroencephalograms (EEGs). This process is expert-biased, tedious, and can delay the diagnosis procedure. Beyond automatically detecting IEDs, there are far fewer studies on automated methods to differentiate epileptic EEGs (potentially without IEDs) from normal EEGs. In addition, the diagnosis of epilepsy based on a single EEG tends to be low. Consequently, there is a strong need for automated systems for EEG interpretation. Traditionally, epilepsy diagnosis relies heavily on IEDs. However, since not all epileptic EEGs exhibit IEDs, it is essential to explore IED-independent EEG measures for epilepsy diagnosis. The main objective is to develop an automated system for detecting epileptic EEGs, both with or without IEDs. In order to detect epileptic EEGs without IEDs, it is crucial to include EEG features in the algorithm that are not directly related to IEDs.Approach.In this study, we explore the background characteristics of interictal EEG for automated and more reliable diagnosis of epilepsy. Specifically, we investigate features based on univariate temporal measures (UTMs), spectral, wavelet, Stockwell, connectivity, and graph metrics of EEGs, besides patient-related information (age and vigilance state). The evaluation is performed on a sizeable cohort of routine scalp EEGs (685 epileptic EEGs and 1229 normal EEGs) from five centers across Singapore, USA, and India.Main results.In comparison with the current literature, we obtained an improved Leave-One-Subject-Out (LOSO) cross-validation (CV) area under the curve (AUC) of 0.871 (Balanced Accuracy (BAC) of 80.9%) with a combination of three features (IED rate, and Daubechies and Morlet wavelets) for the classification of EEGs with IEDs vs. normal EEGs. The IED-independent feature UTM achieved a LOSO CV AUC of 0.809 (BAC of 74.4%). The inclusion of IED-independent features also helps to improve the EEG-level classification of epileptic EEGs with and without IEDs vs. normal EEGs, achieving an AUC of 0.822 (BAC of 77.6%) compared to 0.688 (BAC of 59.6%) for classification only based on the IED rate. Specifically, the addition of IED-independent features improved the BAC by 21% in detecting epileptic EEGs that do not contain IEDs.Significance.These results pave the way towards automated detection of epilepsy. We are one of the first to analyze epileptic EEGs without IEDs, thereby opening up an underexplored option in epilepsy diagnosis.

A machine learning approach to predict progression on active surveillance for prostate cancer.

PURPOSE: Robust prediction of progression on active surveillance (AS) for prostate cancer can allow for risk-adapted protocols. To date, models predicting progression on AS have invariably used traditional statistical approaches. We sought to evaluate whether a machine learning (ML) approach could improve prediction of progression on AS. PATIENTS AND METHODS: We performed a retrospective cohort study of patients diagnosed with very-low or low-risk prostate cancer between 1997 and 2016 and managed with AS at our institution. In the training set, we trained a traditional logistic regression (T-LR) classifier, and alternate ML classifiers (support vector machine, random forest, a fully connected artificial neural network, and ML-LR) to predict grade-progression. We evaluated model performance in the test set. The primary performance metric was the F1 score. RESULTS: Our cohort included 790 patients. With a median follow-up of 6.29 years, 234 developed grade-progression. In descending order, the F1 scores were: support vector machine 0.586 (95% CI 0.579 - 0.591), ML-LR 0.522 (95% CI 0.513 - 0.526), artificial neural network 0.392 (95% CI 0.379 - 0.396), random forest 0.376 (95% CI 0.364 - 0.380), and T-LR 0.182 (95% CI 0.151 - 0.185). All alternate ML models had a significantly higher F1 score than the T-LR model (all p <0.001). CONCLUSION: In our study, ML methods significantly outperformed T-LR in predicting progression on AS for prostate cancer. While our specific models require further validation, we anticipate that a ML approach will help produce robust prediction models that will facilitate individualized risk-stratification in prostate cancer AS.

Time-Frequency Decomposition of Scalp Electroencephalograms Improves Deep Learning-Based Epilepsy Diagnosis.

Epilepsy diagnosis based on Interictal Epileptiform Discharges (IEDs) in scalp electroencephalograms (EEGs) is laborious and often subjective. Therefore, it is necessary to build an effective IED detector and an automatic method to classify IED-free versus IED EEGs. In this study, we evaluate features that may provide reliable IED detection and EEG classification. Specifically, we investigate the IED detector based on convolutional neural network (ConvNet) with different input features (temporal, spectral, and wavelet features). We explore different ConvNet architectures and types, including 1D (one-dimensional) ConvNet, 2D (two-dimensional) ConvNet, and noise injection at various layers. We evaluate the EEG classification performance on five independent datasets. The 1D ConvNet with preprocessed full-frequency EEG signal and frequency bands (delta, theta, alpha, beta) with Gaussian additive noise at the output layer achieved the best IED detection results with a false detection rate of 0.23/min at 90% sensitivity. The EEG classification system obtained a mean EEG classification Leave-One-Institution-Out (LOIO) cross-validation (CV) balanced accuracy (BAC) of 78.1% (area under the curve (AUC) of 0.839) and Leave-One-Subject-Out (LOSO) CV BAC of 79.5% (AUC of 0.856). Since the proposed classification system only takes a few seconds to analyze a 30-min routine EEG, it may help in reducing the human effort required for epilepsy diagnosis.

Postoperative delirium mediates 180-day mortality in orthopaedic trauma patients.

BACKGROUND: Frailty has been associated with increased incidence of postoperative delirium and mortality. We hypothesised that postoperative delirium mediates a clinically significant (≥1%) percentage of the effect of frailty on mortality in older orthopaedic trauma patients. METHODS: This was a single-centre, retrospective observational study including 558 adults 65 yr and older, who presented with an extremity fracture requiring hospitalisation without initial ICU admission. We used causal statistical inference methods to estimate the relationships between frailty, postoperative delirium, and mortality. RESULTS: In the cohort, 180-day mortality rate was 6.5% (36/558). Frail and prefrail patients comprised 23% and 39%, respectively, of the study cohort. Frailty was associated with increased 180 day mortality from 1.4% to 12.2% (11% difference; 95% confidence interval [CI], 8.4-13.6), which translated statistically into an 88.7% (79.9-94.3%) direct effect and an 11.3% (5.7-20.1%) postoperative delirium mediated effect. Prefrailty was also associated with increased 180 day mortality from 1.4% to 4.4% (2.9% difference; 2.4-3.4), which was translated into a 92.5% (83.8-99.9%) direct effect and a 7.5% (0.1-16.2%) postoperative delirium mediated effect. CONCLUSIONS: Frailty is associated with increased postoperative mortality, and delirium might mediate a clinically significant, but small percentage of this effect. Studies should assess whether, in patients with frailty, attempts to mitigate delirium might decrease postoperative mortality.

Automated Adult Epilepsy Diagnostic Tool Based on Interictal Scalp Electroencephalogram Characteristics: A Six-Center Study.

The diagnosis of epilepsy often relies on a reading of routine scalp electroencephalograms (EEGs). Since seizures are highly unlikely to be detected in a routine scalp EEG, the primary diagnosis depends heavily on the visual evaluation of Interictal Epileptiform Discharges (IEDs). This process is tedious, expert-centered, and delays the treatment plan. Consequently, the development of an automated, fast, and reliable epileptic EEG diagnostic system is essential. In this study, we propose a system to classify EEG as epileptic or normal based on multiple modalities extracted from the interictal EEG. The ensemble system consists of three components: a Convolutional Neural Network (CNN)-based IED detector, a Template Matching (TM)-based IED detector, and a spectral feature-based classifier. We evaluate the system on datasets from six centers from the USA, Singapore, and India. The system yields a mean Leave-One-Institution-Out (LOIO) cross-validation (CV) area under curve (AUC) of 0.826 (balanced accuracy (BAC) of 76.1%) and Leave-One-Subject-Out (LOSO) CV AUC of 0.812 (BAC of 74.8%). The LOIO results are found to be similar to the interrater agreement (IRA) reported in the literature for epileptic EEG classification. Moreover, as the proposed system can process routine EEGs in a few seconds, it may aid the clinicians in diagnosing epilepsy efficiently.

Continuous EEG Findings in Autoimmune Encephalitis.

PURPOSE: Autoimmune encephalitis (AE) is a cause of new-onset seizures, including new-onset refractory status epilepticus, yet there have been few studies assessing the EEG signature of AE. METHODS: Multicenter retrospective review of patients diagnosed with AE who underwent continuous EEG monitoring. RESULTS: We identified 64 patients (male, 39%; white, 49%; median age, 44 years); of whom, 43 (67%) were antibody-proven AE patients. Of the patients with confirmed antibody AE, the following were identified: N-methyl-D-aspartate receptor (n = 17, 27%), voltage-gated potassium channel (n = 16, 25%), glutamic acid decarboxylase (n = 6, 9%), and other (n = 4, 6%). The remaining patients were classified as probable antibody-negative AE (n = 11, 17%), definite limbic encephalitis (antibody-negative) (n = 2, 3%), and Hashimoto encephalopathy (n = 8, 13%). Fifty-three percent exhibited electrographic seizures. New-onset refractory status epilepticus was identified in 19% of patients. Sixty-three percent had periodic or rhythmic patterns; of which, 38% had plus modifiers. Generalized rhythmic delta activity was identified in 33% of patients. Generalized rhythmic delta activity and generalized rhythmic delta activity plus fast activity were more common in anti-N-methyl-D-aspartate AE (P = 0.0001 and 0.0003, respectively). No other periodic or rhythmic patterns exhibited AE subtype association. Forty-two percent had good outcome on discharge. Periodic or rhythmic patterns, seizures, and new-onset refractory status epilepticus conferred an increased risk of poor outcome (OR, 6.4; P = 0.0012; OR, 3; P = 0.0372; OR, 12.3; P = 0.02, respectively). CONCLUSION: Our study confirms a signature EEG pattern in anti-N-methyl-D-aspartate AE, termed extreme delta brush, identified as generalized rhythmic delta activity plus fast activity in our study. We found no other pattern association with other AE subtypes. We also found a high incidence of seizures among patients with AE. Finally, periodic or rhythmic patterns, seizures, and new-onset refractory status epilepticus conferred an increased risk of poor outcome regardless of AE subtype.

Deep Learning for Interictal Epileptiform Spike Detection from scalp EEG frequency sub bands.

Epilepsy diagnosis through visual examination of interictal epileptiform discharges (IEDs) in scalp electroencephalogram (EEG) signals is a challenging problem. Deep learning methods can be an automated way to perform this task. In this work, we present a new approach based on convolutional neural network (CNN) to detect IEDs from EEGs automatically. The input to CNN is a combination of raw EEG and frequency sub-bands, namely delta, theta, alpha and, beta arranged as a vector for one-dimensional (1D) CNN or matrix for two-dimensional (2D) CNN. The proposed method is evaluated on 554 scalp EEGs. The database consists of 18,164 IEDs marked by two neurologists. Five-fold cross-validation was performed to assess the IED detectors. The resulting 1D CNN based IED detector with multiple sub-bands achieved a false positive rate per minute of 0.23 and a precision of 0.79 at 90% sensitivity. Further, the proposed system is evaluated on datasets from three other clinics, and the features extracted from CNN outputs could significantly discriminate (p-values <; 0.05) the EEGs with and without IEDs. We have proposed an optimized method with better performance than the literature that could aid clinicians to diagnose epilepsy expeditiously, and thereby devise proper treatment.

Automated Detection of Interictal Epileptiform Discharges from Scalp Electroencephalograms by Convolutional Neural Networks.

Visual evaluation of electroencephalogram (EEG) for Interictal Epileptiform Discharges (IEDs) as distinctive biomarkers of epilepsy has various limitations, including time-consuming reviews, steep learning curves, interobserver variability, and the need for specialized experts. The development of an automated IED detector is necessary to provide a faster and reliable diagnosis of epilepsy. In this paper, we propose an automated IED detector based on Convolutional Neural Networks (CNNs). We have evaluated the proposed IED detector on a sizable database of 554 scalp EEG recordings (84 epileptic patients and 461 nonepileptic subjects) recorded at Massachusetts General Hospital (MGH), Boston. The proposed CNN IED detector has achieved superior performance in comparison with conventional methods with a mean cross-validation area under the precision-recall curve (AUPRC) of 0.838[Formula: see text]±[Formula: see text]0.040 and false detection rate of 0.2[Formula: see text]±[Formula: see text]0.11 per minute for a sensitivity of 80%. We demonstrated the proposed system to be noninferior to 30 neurologists on a dataset from the Medical University of South Carolina (MUSC). Further, we clinically validated the system at National University Hospital (NUH), Singapore, with an agreement accuracy of 81.41% with a clinical expert. Moreover, the proposed system can be applied to EEG recordings with any arbitrary number of channels.

Development and Prospective Validation of a Transparent Deep Learning Algorithm for Predicting Need for Mechanical Ventilation.

IMPORTANCE: Objective and early identification of hospitalized patients, and particularly those with novel coronavirus disease 2019 (COVID-19), who may require mechanical ventilation is of great importance and may aid in delivering timely treatment. OBJECTIVE: To develop, externally validate and prospectively test a transparent deep learning algorithm for predicting 24 hours in advance the need for mechanical ventilation in hospitalized patients and those with COVID-19. DESIGN: Observational cohort study SETTING: Two academic medical centers from January 01, 2016 to December 31, 2019 (Retrospective cohorts) and February 10, 2020 to May 4, 2020 (Prospective cohorts). PARTICIPANTS: Over 31,000 admissions to the intensive care units (ICUs) at two hospitals. Additionally, 777 patients with COVID-19 patients were used for prospective validation. Patients who were placed on mechanical ventilation within four hours of their admission were excluded. MAIN OUTCOME(S) and MEASURE(S): Electronic health record (EHR) data were extracted on an hourly basis, and a set of 40 features were calculated and passed to an interpretable deep-learning algorithm to predict the future need for mechanical ventilation 24 hours in advance. Additionally, commonly used clinical criteria (based on heart rate, oxygen saturation, respiratory rate, FiO2 and pH) was used to assess future need for mechanical ventilation. Performance of the algorithms were evaluated using the area under receiver-operating characteristic curve (AUC), sensitivity, specificity and positive predictive value. RESULTS: After applying exclusion criteria, the external validation cohort included 3,888 general ICU and 402 COVID-19 patients. The performance of the model (AUC) with a 24-hour prediction horizon at the validation site was 0.882 for the general ICU population and 0.918 for patients with COVID-19. In comparison, commonly used clinical criteria and the ROX score achieved AUCs in the range of 0.773 - 0.782 and 0.768 - 0.810 for the general ICU population and patients with COVID-19, respectively. CONCLUSIONS AND RELEVANCE: A generalizable and transparent deep-learning algorithm improves on traditional clinical criteria to predict the need for mechanical ventilation in hospitalized patients, including those with COVID-19. Such an algorithm may help clinicians with optimizing timing of tracheal intubation, better allocation of mechanical ventilation resources and staff, and improve patient care.

Development and Validation of Forecasting Next Reported Seizure Using e-Diaries.

OBJECTIVE: There are no validated methods for predicting the timing of seizures. Using machine learning, we sought to forecast 24-hour risk of self-reported seizure from e-diaries. METHODS: Data from 5,419 patients on SeizureTracker.com (including seizure count, type, and duration) were split into training (3,806 patients/1,665,215 patient-days) and testing (1,613 patients/549,588 patient-days) sets with no overlapping patients. An artificial intelligence (AI) program, consisting of recurrent networks followed by a multilayer perceptron ("deep learning" model), was trained to produce risk forecasts. Forecasts were made from a sliding window of 3-month diary history for each day of each patient's diary. After training, the model parameters were held constant and the testing set was scored. A rate-matched random (RMR) forecast was compared to the AI. Comparisons were made using the area under the receiver operating characteristic curve (AUC), a measure of binary discrimination performance, and the Brier score, a measure of forecast calibration. The Brier skill score (BSS) measured the improvement of the AI Brier score compared to the benchmark RMR Brier score. Confidence intervals (CIs) on performance statistics were obtained via bootstrapping. RESULTS: The AUC was 0.86 (95% CI = 0.85-0.88) for AI and 0.83 (95% CI = 0.81-0.85) for RMR, favoring AI (p < 0.001). Overall (all patients combined), BSS was 0.27 (95% CI = 0.23-0.31), also favoring AI (p < 0.001). INTERPRETATION: The AI produced a valid forecast superior to a chance forecaster, and provided meaningful forecasts in the majority of patients. Future studies will be needed to quantify the clinical value of these forecasts for patients. ANN NEUROL 2020;88:588-595.

Brain Monitoring of Sedation in the Intensive Care Unit Using a Recurrent Neural Network.

Over and under-sedation are common in critically ill patients admitted to the Intensive Care Unit. Clinical assessments provide limited time resolution and are based on behavior rather than the brain itself. Existing brain monitors have been developed primarily for non-ICU settings. Here, we use a clinical dataset from 154 ICU patients in whom the Richmond Agitation-Sedation Score is assessed about every 2 hours. We develop a recurrent neural network (RNN) model to discriminate between deep vs. no sedation, trained end-to-end from raw EEG spectrograms without any feature extraction. We obtain an average area under the ROC of 0.8 on 10-fold cross validation across patients. Our RNN is able to provide reliable estimates of sedation levels consistently better compared to a feed-forward model with simple smoothing. Decomposing the prediction error in terms of sedatives reveals that patient-specific calibration for sedatives is expected to further improve sedation monitoring.

Neuroimaging Correlates of Periodic Discharges.

Despite being first described over 50 years ago, periodic discharges continue to generate controversy as to whether they are always, sometimes, or never "ictal." Investigators and clinicians have proposed adjunctive markers to help clarify this distinction-in particular measures of perfusion and metabolism. Here, we review the growing number of neuroimaging studies using Fluorodeoxyglucose-PET, MRI diffusion, Magnetic resonance perfusion, Single Photon Emission Computed Tomography, and Magnetoencepgalography to gain further insight into the physiology and clinical significance of periodic discharges. To date, however, no definitive consensus exists regarding the features of periodic discharges that warrant treatment intensification. However, an emerging consilience among neuroimaging modalities suggests that periodic discharges can induce a hyperexcitatory state with associated hypermetabolism and hyperperfusion, which may result in local metabolic failure.

New-onset refractory status epilepticus: Etiology, clinical features, and outcome.

OBJECTIVES: The aims of this study were to determine the etiology, clinical features, and predictors of outcome of new-onset refractory status epilepticus. METHODS: Retrospective review of patients with refractory status epilepticus without etiology identified within 48 hours of admission between January 1, 2008, and December 31, 2013, in 13 academic medical centers. The primary outcome measure was poor functional outcome at discharge (defined as a score >3 on the modified Rankin Scale). RESULTS: Of 130 cases, 67 (52%) remained cryptogenic. The most common identified etiologies were autoimmune (19%) and paraneoplastic (18%) encephalitis. Full data were available in 125 cases (62 cryptogenic). Poor outcome occurred in 77 of 125 cases (62%), and 28 (22%) died. Predictors of poor outcome included duration of status epilepticus, use of anesthetics, and medical complications. Among the 63 patients with available follow-up data (median 9 months), functional status improved in 36 (57%); 79% had good or fair outcome at last follow-up, but epilepsy developed in 37% with most survivors (92%) remaining on antiseizure medications. Immune therapies were used less frequently in cryptogenic cases, despite a comparable prevalence of inflammatory CSF changes. CONCLUSIONS: Autoimmune encephalitis is the most commonly identified cause of new-onset refractory status epilepticus, but half remain cryptogenic. Outcome at discharge is poor but improves during follow-up. Epilepsy develops in most cases. The role of anesthetics and immune therapies warrants further investigation.

Addressing Big Data Time Series: Mining Trillions of Time Series Subsequences Under Dynamic Time Warping.

Most time series data mining algorithms use similarity search as a core subroutine, and thus the time taken for similarity search is the bottleneck for virtually all time series data mining algorithms, including classification, clustering, motif discovery, anomaly detection, and so on. The difficulty of scaling a search to large datasets explains to a great extent why most academic work on time series data mining has plateaued at considering a few millions of time series objects, while much of industry and science sits on billions of time series objects waiting to be explored. In this work we show that by using a combination of four novel ideas we can search and mine massive time series for the first time. We demonstrate the following unintuitive fact: in large datasets we can exactly search under Dynamic Time Warping (DTW) much more quickly than the current state-of-the-art Euclidean distance search algorithms. We demonstrate our work on the largest set of time series experiments ever attempted. In particular, the largest dataset we consider is larger than the combined size of all of the time series datasets considered in all data mining papers ever published. We explain how our ideas allow us to solve higher-level time series data mining problems such as motif discovery and clustering at scales that would otherwise be untenable. Moreover, we show how our ideas allow us to efficiently support the uniform scaling distance measure, a measure whose utility seems to be underappreciated, but which we demonstrate here. In addition to mining massive datasets with up to one trillion datapoints, we will show that our ideas also have implications for real-time monitoring of data streams, allowing us to handle much faster arrival rates and/or use cheaper and lower powered devices than are currently possible.