Improving Neurology Clinical Care With Natural Language Processing Tools.

The integration of natural language processing (NLP) tools into neurology workflows has the potential to significantly enhance clinical care. However, it is important to address the limitations and risks associated with integrating this new technology. Recent advances in transformer-based NLP algorithms (e.g., GPT, BERT) could augment neurology clinical care by summarizing patient health information, suggesting care options, and assisting research involving large datasets. However, these NLP platforms have potential risks including fabricated facts and data security and substantial barriers for implementation. Although these risks and barriers need to be considered, the benefits for providers, patients, and communities are substantial. With these systems achieving greater functionality and the pace of medical need increasing, integrating these tools into clinical care may prove not only beneficial but necessary. Further investigation is needed to design implementation strategies, mitigate risks, and overcome barriers.

Automated detection of immune effector cell-associated neurotoxicity syndrome via quantitative EEG.

OBJECTIVE: To develop an automated, physiologic metric of immune effector cell-associated neurotoxicity syndrome among patients undergoing chimeric antigen receptor-T cell therapy. METHODS: We conducted a retrospective observational cohort study from 2016 to 2020 at two tertiary care centers among patients receiving chimeric antigen receptor-T cell therapy with a CD19 or B-cell maturation antigen ligand. We determined the daily neurotoxicity grade for each patient during EEG monitoring via chart review and extracted clinical variables and outcomes from the electronic health records. Using quantitative EEG features, we developed a machine learning model to detect the presence and severity of neurotoxicity, known as the EEG immune effector cell-associated neurotoxicity syndrome score. RESULTS: The EEG immune effector cell-associated neurotoxicity syndrome score significantly correlated with the grade of neurotoxicity with a median Spearman's R2 of 0.69 (95% CI of 0.59-0.77). The mean area under receiving operator curve was greater than 0.85 for each binary discrimination level. The score also showed significant correlations with maximum ferritin (R2 0.24, p = 0.008), minimum platelets (R2 -0.29, p = 0.001), and dexamethasone usage (R2 0.42, p < 0.0001). The score significantly correlated with duration of neurotoxicity (R2 0.31, p < 0.0001). INTERPRETATION: The EEG immune effector cell-associated neurotoxicity syndrome score possesses high criterion, construct, and predictive validity, which substantiates its use as a physiologic method to detect the presence and severity of neurotoxicity among patients undergoing chimeric antigen receptor T-cell therapy.

Effects of epileptiform activity on discharge outcome in critically ill patients in the USA: a retrospective cross-sectional study.

BACKGROUND: Epileptiform activity is associated with worse patient outcomes, including increased risk of disability and death. However, the effect of epileptiform activity on neurological outcome is confounded by the feedback between treatment with antiseizure medications and epileptiform activity burden. We aimed to quantify the heterogeneous effects of epileptiform activity with an interpretability-centred approach. METHODS: We did a retrospective, cross-sectional study of patients in the intensive care unit who were admitted to Massachusetts General Hospital (Boston, MA, USA). Participants were aged 18 years or older and had electrographic epileptiform activity identified by a clinical neurophysiologist or epileptologist. The outcome was the dichotomised modified Rankin Scale (mRS) at discharge and the exposure was epileptiform activity burden defined as mean or maximum proportion of time spent with epileptiform activity in 6 h windows in the first 24 h of electroencephalography. We estimated the change in discharge mRS if everyone in the dataset had experienced a specific epileptiform activity burden and were untreated. We combined pharmacological modelling with an interpretable matching method to account for confounding and epileptiform activity-antiseizure medication feedback. The quality of the matched groups was validated by the neurologists. FINDINGS: Between Dec 1, 2011, and Oct 14, 2017, 1514 patients were admitted to Massachusetts General Hospital intensive care unit, 995 (66%) of whom were included in the analysis. Compared with patients with a maximum epileptiform activity of 0 to less than 25%, patients with a maximum epileptiform activity burden of 75% or more when untreated had a mean 22·27% (SD 0·92) increased chance of a poor outcome (severe disability or death). Moderate but long-lasting epileptiform activity (mean epileptiform activity burden 2% to <10%) increased the risk of a poor outcome by mean 13·52% (SD 1·93). The effect sizes were heterogeneous depending on preadmission profile-eg, patients with hypoxic-ischaemic encephalopathy or acquired brain injury were more adversely affected compared with patients without these conditions. INTERPRETATION: Our results suggest that interventions should put a higher priority on patients with an average epileptiform activity burden 10% or greater, and treatment should be more conservative when maximum epileptiform activity burden is low. Treatment should also be tailored to individual preadmission profiles because the potential for epileptiform activity to cause harm depends on age, medical history, and reason for admission. FUNDING: National Institutes of Health and National Science Foundation.

Association of Epileptiform Activity With Outcomes in Toxic-Metabolic Encephalopathy.

The clinical significance of epileptiform abnormalities (EAs) specific to toxic-metabolic encephalopathy (TME) is unknown. OBJECTIVES: To quantify EA burden in patients with TME and its association with neurologic outcomes. DESIGN SETTING AND PARTICIPANT: This is a retrospective study. A cohort of patients with TME and EA (positive) were age, Sequential Organ Failure Assessment Score, Acute Physiology and Chronic Health Evaluation II (APACHE-II) score matched to a cohort of TME patients without EA (control). Univariate analysis compared EA-positive patients against controls. Multivariable logistical regression adjusting for underlying disease etiology was performed to examine the relationship between EA burden and probability of poor neurologic outcome (modified Rankin Score [mRS] 4-6) at discharge. Consecutive admissions to inpatient floors or ICUs that underwent continuous electroencephalography (cEEG) monitoring at a single center between 2012 and 2019. Inclusion criteria were 1) patients with TME diagnosis, 2) age greater than 18 years, and 3) greater than or equal to 16 hours of cEEG. Patients with acute brain injury and cardiac arrest were excluded. MAIN OUTCOMES AND MEASURES: Poor neurologic outcome defined by mRS (mRS 4-6). RESULTS: One hundred sixteen patients were included, 58 with EA and 58 controls without EA, where matching was performed on age and APACHE-II score. The median age was 66 (Q1-Q3, 57-75) and median APACHE II score was 18 (Q1-Q3, 13-22). Overall cohort discharge mortality was 22% and 70% had a poor neurologic outcome. Peak EA burden was defined as the 12-hour window of recording with the highest prevalence of EAs. In multivariable analysis adjusted for Charlson Comorbidity Index and primary diagnosis, presence of EAs was associated with poor outcome (odds ratio 3.89; CI [1.05-14.2], p = 0.041). Increase in peak EA burden from 0% to 100% increased probability of poor discharge neurologic outcome by 30%. CONCLUSIONS AND RELEVANCE: Increasing burden of EA is associated with worse discharge outcomes in patients with TME. Future studies are needed to determine whether short-term treatment with anti-seizure medications while medically treating the underlying metabolic derangement improves outcomes.

Toward carbon neutrality: Projecting a desert-based photovoltaic power network circumnavigating the globe.

Carbon, the human's most reliable fuel type in the past, must be neutralized in this century toward the Paris Agreement temperature goals. Solar power is widely believed a key fossil fuel substitute but suffers from the needs of large space occupation and huge energy storage for peak shaving. Here, we propose a solar network circumnavigating the globe to connecting large-scale desert photovoltaics among continents. By evaluating the generation potential of desert photovoltaic plants on each continent (taking dust accumulation into account) and the hourly maximum transmission potential that each inhabited continent can receive (taking transmission loss into account), we find that the current total annual human demand for electricity will be more than met by this solar network. The local imbalanced diurnal generation of photovoltaic energy can be made up by transcontinental power transmission from other power stations in the network to meet the hourly electricity demand. We also find that laying solar panels over a large space may darken the Earth's surface, but this albedo warming effect is orders of magnitude lower than that of CO2 released from thermal power plants. From practical needs and ecological effects, this powerful and stable power network with lower climate perturbability could potentially help to phase out global carbon emissions in the 21st century.

Identification of patients with epilepsy using automated electronic health records phenotyping.

OBJECTIVE: Unstructured data present in electronic health records (EHR) are a rich source of medical information; however, their abstraction is labor intensive. Automated EHR phenotyping (AEP) can reduce the need for manual chart review. We present an AEP model that is designed to automatically identify patients diagnosed with epilepsy. METHODS: The ground truth for model training and evaluation was captured from a combination of structured questionnaires filled out by physicians for a subset of patients and manual chart review using customized software. Modeling features included indicators of the presence of keywords and phrases in unstructured clinical notes, prescriptions for antiseizure medications (ASMs), International Classification of Diseases (ICD) codes for seizures and epilepsy, number of ASMs and epilepsy-related ICD codes, age, and sex. Data were randomly divided into training (70%) and hold-out testing (30%) sets, with distinct patients in each set. We trained regularized logistic regression and an extreme gradient boosting models. Model performance was measured using area under the receiver operating curve (AUROC) and area under the precision-recall curve (AUPRC), with 95% confidence intervals (CI) estimated via bootstrapping. RESULTS: Our study cohort included 3903 adults drawn from outpatient departments of nine hospitals between February 2015 and June 2022 (mean age = 47 ± 18 years, 57% women, 82% White, 84% non-Hispanic, 70% with epilepsy). The final models included 285 features, including 246 keywords and phrases captured from 8415 encounters. Both models achieved AUROC and AUPRC of 1 (95% CI = .99-1.00) in the hold-out testing set. SIGNIFICANCE: A machine learning-based AEP approach accurately identifies patients with epilepsy from notes, ICD codes, and ASMs. This model can enable large-scale epilepsy research using EHR databases.

Classification of neurologic outcomes from medical notes using natural language processing.

Neurologic disability level at hospital discharge is an important outcome in many clinical research studies. Outside of clinical trials, neurologic outcomes must typically be extracted by labor intensive manual review of clinical notes in the electronic health record (EHR). To overcome this challenge, we set out to develop a natural language processing (NLP) approach that automatically reads clinical notes to determine neurologic outcomes, to make it possible to conduct larger scale neurologic outcomes studies. We obtained 7314 notes from 3632 patients hospitalized at two large Boston hospitals between January 2012 and June 2020, including discharge summaries (3485), occupational therapy (1472) and physical therapy (2357) notes. Fourteen clinical experts reviewed notes to assign scores on the Glasgow Outcome Scale (GOS) with 4 classes, namely 'good recovery', 'moderate disability', 'severe disability', and 'death' and on the Modified Rankin Scale (mRS), with 7 classes, namely 'no symptoms', 'no significant disability', 'slight disability', 'moderate disability', 'moderately severe disability', 'severe disability', and 'death'. For 428 patients' notes, 2 experts scored the cases generating interrater reliability estimates for GOS and mRS. After preprocessing and extracting features from the notes, we trained a multiclass logistic regression model using LASSO regularization and 5-fold cross validation for hyperparameter tuning. The model performed well on the test set, achieving a micro average area under the receiver operating characteristic and F-score of 0.94 (95% CI 0.93-0.95) and 0.77 (0.75-0.80) for GOS, and 0.90 (0.89-0.91) and 0.59 (0.57-0.62) for mRS, respectively. Our work demonstrates that an NLP algorithm can accurately assign neurologic outcomes based on free text clinical notes. This algorithm increases the scale of research on neurological outcomes that is possible with EHR data.

Development of Expert-Level Classification of Seizures and Rhythmic and Periodic Patterns During EEG Interpretation.

BACKGROUND AND OBJECTIVES: Seizures (SZs) and other SZ-like patterns of brain activity can harm the brain and contribute to in-hospital death, particularly when prolonged. However, experts qualified to interpret 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 SZs and other SZ-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 SZs and SZ-like events, known as "ictal-interictal-injury continuum" (IIIC) patterns on EEG, including SZs, 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 data sets 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 by the calibration index and by the percentage of experts whose operating points were below the model's receiver operating characteristic curves (ROCs) and precision recall curves (PRCs) 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%, and 40%; PRC: 50%, 35%, 50%, 90%, 70%, and 45%; and calibration: 95%, 100%, 95%, 100%, 100%, and 80%, respectively. DISCUSSION: SPaRCNet is the first algorithm to match expert performance in detecting SZs and other SZ-like events in a representative sample of EEGs. With further development, SPaRCNet may thus be a valuable tool for an expedited review of EEGs. CLASSIFICATION OF EVIDENCE: This study provides Class II evidence that among patients with epilepsy or critical illness undergoing EEG monitoring, SPaRCNet can differentiate (IIIC) patterns from non-IIIC events and expert neurophysiologists.

Quantitative epileptiform burden and electroencephalography background features predict post-traumatic epilepsy.

BACKGROUND: Post-traumatic epilepsy (PTE) is a severe complication of traumatic brain injury (TBI). Electroencephalography aids early post-traumatic seizure diagnosis, but its optimal utility for PTE prediction remains unknown. We aim to evaluate the contribution of quantitative electroencephalograms to predict first-year PTE (PTE1). METHODS: We performed a multicentre, retrospective case-control study of patients with TBI. 63 PTE1 patients were matched with 63 non-PTE1 patients by admission Glasgow Coma Scale score, age and sex. We evaluated the association of quantitative electroencephalography features with PTE1 using logistic regressions and examined their predictive value relative to TBI mechanism and CT abnormalities. RESULTS: In the matched cohort (n=126), greater epileptiform burden, suppression burden and beta variability were associated with 4.6 times higher PTE1 risk based on multivariable logistic regression analysis (area under the receiver operating characteristic curve, AUC (95% CI) 0.69 (0.60 to 0.78)). Among 116 (92%) patients with available CT reports, adding quantitative electroencephalography features to a combined mechanism and CT model improved performance (AUC (95% CI), 0.71 (0.61 to 0.80) vs 0.61 (0.51 to 0.72)). CONCLUSIONS: Epileptiform and spectral characteristics enhance covariates identified on TBI admission and CT abnormalities in PTE1 prediction. Future trials should incorporate quantitative electroencephalography features to validate this enhancement of PTE risk stratification models.

Interrater Reliability of Expert Electroencephalographers Identifying Seizures and Rhythmic and Periodic Patterns in EEGs.

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). Previous interrater 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 (LPDs)," "generalized periodic discharges (GPDs)," "lateralized rhythmic delta activity (LRDA)," "generalized rhythmic delta activity (GRDA)," or "other." EEGs were performed for clinical indications at Massachusetts General Hospital between 2006 and 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 women, and the median (IQR) age was 55 (41) years. 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 experts' false-positive vs true-positive characteristics (median [range] of variance explained ([Formula: see text]): 95 [93, 98]%) and for most variation in experts' precision vs sensitivity characteristics ([Formula: see text]: 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 an independent expert review reliably identifies ictal-interictal injury continuum patterns on EEG compared with expert consensus.

EEG-based grading of immune effector cell-associated neurotoxicity syndrome.

CAR-T cell therapy is an effective cancer therapy for multiple refractory/relapsed hematologic malignancies but is associated with substantial toxicity, including Immune Effector Cell Associated Neurotoxicity Syndrome (ICANS). Improved detection and assessment of ICANS could improve management and allow greater utilization of CAR-T cell therapy, however, an objective, specific biomarker has not been identified. We hypothesized that the severity of ICANS can be quantified based on patterns of abnormal brain activity seen in electroencephalography (EEG) signals. We conducted a retrospective observational study of 120 CAR-T cell therapy patients who had received EEG monitoring. We determined a daily ICANS grade for each patient through chart review. We used visually assessed EEG features and machine learning techniques to develop the Visual EEG-Immune Effector Cell Associated Neurotoxicity Syndrome (VE-ICANS) score and assessed the association between VE-ICANS and ICANS. We also used it to determine the significance and relative importance of the EEG features. We developed the Visual EEG-ICANS (VE-ICANS) grading scale, a grading scale with a physiological basis that has a strong correlation to ICANS severity (R = 0.58 [0.47-0.66]) and excellent discrimination measured via area under the receiver operator curve (AUC = 0.91 for ICANS ≥ 2). This scale shows promise as a biomarker for ICANS which could help to improve clinical care through greater accuracy in assessing ICANS severity.

Evaluation of the Street Canyon Level Air Pollution Distribution Pattern in a Typical City Block in Baoding, China.

Urban traffic pollution, which is strongly influenced by the complex urban morphology, has posed a great threat to human health. In this study, we performed a high-resolution simulation of traffic pollution in a typical city block in Baoding, China, based on the Parallelized Large-eddy simulation Model (PALM), to examine the distribution patterns of traffic-related pollutants and explore their relationship with urban morphology. Based on the model results, we conducted a multi-linear regression (MLR) analysis and found that the distribution of air pollutants inside the city block was dominated by both traffic emissions and urban morphology, which explained about 70% of the total variance in spatial distribution of air pollutants. Excluding the contribution of emissions, over 50% of the total variance can still be explained by the urban morphology. Among these urban morphological factors, the key factors determining the spatial distribution of air pollution are "Distance from the road" (DR), "Building Coverage Ratio" (BCR) and "Aspect Ratio" (H/W) of the street canyon. Specifically, urban areas with lower Aspect Ratio, lower BCR and larger DR are less affected by traffic pollution. Compiling these individual factors, we developed a complex Urban Morphology Pollution Index (UMPI). Each unit increase in UMPI is associated with a one percent increase of nearby traffic pollution contribution. This index can help urban planners to semi-quantitatively evaluate building groups which tend to trap or ventilate traffic pollution and thus help to reduce human exposure to street canyon level pollution through either traffic emission control or urban morphology amelioration.

Identifying Patients With Delirium Based on Unstructured Clinical Notes: Observational Study.

BACKGROUND: Delirium in hospitalized patients is a syndrome of acute brain dysfunction. Diagnostic (International Classification of Diseases [ICD]) codes are often used in studies using electronic health records (EHRs), but they are inaccurate. OBJECTIVE: We sought to develop a more accurate method using natural language processing (NLP) to detect delirium episodes on the basis of unstructured clinical notes. METHODS: We collected 1.5 million notes from >10,000 patients from among 9 hospitals. Seven experts iteratively labeled 200,471 sentences. Using these, we trained three NLP classifiers: Support Vector Machine, Recurrent Neural Networks, and Transformer. Testing was performed using an external data set. We also evaluated associations with delirium billing (ICD) codes, medications, orders for restraints and sitters, direct assessments (Confusion Assessment Method [CAM] scores), and in-hospital mortality. F1 scores, confusion matrices, and areas under the receiver operating characteristic curve (AUCs) were used to compare NLP models. We used the φ coefficient to measure associations with other delirium indicators. RESULTS: The transformer NLP performed best on the following parameters: micro F1=0.978, macro F1=0.918, positive AUC=0.984, and negative AUC=0.992. NLP detections exhibited higher correlations (φ) than ICD codes with deliriogenic medications (0.194 vs 0.073 for ICD codes), restraints and sitter orders (0.358 vs 0.177), mortality (0.216 vs 0.000), and CAM scores (0.256 vs -0.028). CONCLUSIONS: Clinical notes are an attractive alternative to ICD codes for EHR delirium studies but require automated methods. Our NLP model detects delirium with high accuracy, similar to manual chart review. Our NLP approach can provide more accurate determination of delirium for large-scale EHR-based studies regarding delirium, quality improvement, and clinical trails.

Measuring expertise in identifying interictal epileptiform discharges.

OBJECTIVE: Interictal epileptiform discharges on EEG are integral to diagnosing epilepsy. However, EEGs are interpreted by readers with and without specialty training, and there is no accepted method to assess skill in interpretation. We aimed to develop a test to quantify IED recognition skills. METHODS: A total of 13,262 candidate IEDs were selected from EEGs and scored by eight fellowship-trained reviewers to establish a gold standard. An online test was developed to assess how well readers with different training levels could distinguish candidate waveforms. Sensitivity, false positive rate and calibration were calculated for each reader. A simple mathematical model was developed to estimate each reader's skill and threshold in identifying an IED, and to develop receiver operating characteristics curves for each reader. We investigated the number of IEDs needed to measure skill level with acceptable precision. RESULTS: Twenty-nine raters completed the test; nine experts, seven experienced non-experts and thirteen novices. Median calibration errors for experts, experienced non-experts and novices were -0.056, 0.012, 0.046; median sensitivities were 0.800, 0.811, 0.715; and median false positive rates were 0.177, 0.272, 0.396, respectively. The number of test questions needed to measure those scores was 549. Our analysis identified that novices had a higher noise level (uncertainty) compared to experienced non-experts and experts. Using calculated noise and threshold levels, receiver operating curves were created, showing increasing median area under the curve from novices (0.735), to experienced non-experts (0.852) and experts (0.891). SIGNIFICANCE: Expert and non-expert readers can be distinguished based on ability to identify IEDs. This type of assessment could also be used to identify and correct differences in thresholds in identifying IEDs.

Predicting neurological outcome in comatose patients after cardiac arrest with multiscale deep neural networks.

OBJECTIVE: Electroencephalography (EEG) is an important tool for neurological outcome prediction after cardiac arrest. However, the complexity of continuous EEG data limits timely and accurate interpretation by clinicians. We develop a deep neural network (DNN) model to leverage complex EEG trends for early and accurate assessment of cardiac arrest coma recovery likelihood. METHODS: We developed a multiscale DNN combining convolutional neural networks (CNN) and recurrent neural networks (long short-term memory [LSTM]) using EEG and demographic information (age, gender, shockable rhythm) from a multicenter cohort of 1,038 cardiac arrest patients. The CNN learns EEG feature representations while the multiscale LSTM captures short-term and long-term EEG dynamics on multiple time scales. Poor outcome is defined as a Cerebral Performance Category (CPC) score of 3-5 and good outcome as CPC score 1-2 at 3-6 months after cardiac arrest. Performance is evaluated using area under the receiver operating characteristic curve (AUC) and calibration error. RESULTS: Model performance increased with EEG duration, with AUC increasing from 0.83 (95% Confidence Interval [CI] 0.79-0.87 at 12h to 0.91 (95%CI 0.88-0.93) at 66h. Sensitivity of good and poor outcome prediction was 77% and 75% at a specificity of 90%, respectively. Sensitivity of poor outcome was 50% at a specificity of 99%. Predicted probability was well matched to the observation frequency of poor outcomes, with a calibration error of 0.11 [0.09-0.14]. CONCLUSIONS: These results demonstrate that incorporating EEG evolution over time improves the accuracy of neurologic outcome prediction for patients with coma after cardiac arrest.

Automated Annotation of Epileptiform Burden and Its Association with Outcomes.

OBJECTIVE: This study was undertaken to determine the dose-response relation between epileptiform activity burden and outcomes in acutely ill patients. METHODS: A single center retrospective analysis was made of 1,967 neurologic, medical, and surgical patients who underwent >16 hours of continuous electroencephalography (EEG) between 2011 and 2017. We developed an artificial intelligence algorithm to annotate 11.02 terabytes of EEG and quantify epileptiform activity burden within 72 hours of recording. We evaluated burden (1) in the first 24 hours of recording, (2) in the 12-hours epoch with highest burden (peak burden), and (3) cumulatively through the first 72 hours of monitoring. Machine learning was applied to estimate the effect of epileptiform burden on outcome. Outcome measure was discharge modified Rankin Scale, dichotomized as good (0-4) versus poor (5-6). RESULTS: Peak epileptiform burden was independently associated with poor outcomes (p < 0.0001). Other independent associations included age, Acute Physiology and Chronic Health Evaluation II score, seizure on presentation, and diagnosis of hypoxic-ischemic encephalopathy. Model calibration error was calculated across 3 strata based on the time interval between last EEG measurement (up to 72 hours of monitoring) and discharge: (1) <5 days between last measurement and discharge, 0.0941 (95% confidence interval [CI] = 0.0706-0.1191); 5 to 10 days between last measurement and discharge, 0.0946 (95% CI = 0.0631-0.1290); >10 days between last measurement and discharge, 0.0998 (95% CI = 0.0698-0.1335). After adjusting for covariates, increase in peak epileptiform activity burden from 0 to 100% increased the probability of poor outcome by 35%. INTERPRETATION: Automated measurement of peak epileptiform activity burden affords a convenient, consistent, and quantifiable target for future multicenter randomized trials investigating whether suppressing epileptiform activity improves outcomes. ANN NEUROL 2021;90:300-311.

Prolonged Intubation in Patients With Prior Cerebrovascular Disease and COVID-19.

Objectives: Patients with comorbidities are at increased risk for poor outcomes in COVID-19, yet data on patients with prior neurological disease remains limited. Our objective was to determine the odds of critical illness and duration of mechanical ventilation in patients with prior cerebrovascular disease and COVID-19. Methods: A observational study of 1,128 consecutive adult patients admitted to an academic center in Boston, Massachusetts, and diagnosed with laboratory-confirmed COVID-19. We tested the association between prior cerebrovascular disease and critical illness, defined as mechanical ventilation (MV) or death by day 28, using logistic regression with inverse probability weighting of the propensity score. Among intubated patients, we estimated the cumulative incidence of successful extubation without death over 45 days using competing risk analysis. Results: Of the 1,128 adults with COVID-19, 350 (36%) were critically ill by day 28. The median age of patients was 59 years (SD: 18 years) and 640 (57%) were men. As of June 2nd, 2020, 127 (11%) patients had died. A total of 177 patients (16%) had a prior cerebrovascular disease. Prior cerebrovascular disease was significantly associated with critical illness (OR = 1.54, 95% CI = 1.14-2.07), lower rate of successful extubation (cause-specific HR = 0.57, 95% CI = 0.33-0.98), and increased duration of intubation (restricted mean time difference = 4.02 days, 95% CI = 0.34-10.92) compared to patients without cerebrovascular disease. Interpretation: Prior cerebrovascular disease adversely affects COVID-19 outcomes in hospitalized patients. Further study is required to determine if this subpopulation requires closer monitoring for disease progression during COVID-19.

Classification of the Disposition of Patients Hospitalized with COVID-19: Reading Discharge Summaries Using Natural Language Processing.

BACKGROUND: Medical notes are a rich source of patient data; however, the nature of unstructured text has largely precluded the use of these data for large retrospective analyses. Transforming clinical text into structured data can enable large-scale research studies with electronic health records (EHR) data. Natural language processing (NLP) can be used for text information retrieval, reducing the need for labor-intensive chart review. Here we present an application of NLP to large-scale analysis of medical records at 2 large hospitals for patients hospitalized with COVID-19. OBJECTIVE: Our study goal was to develop an NLP pipeline to classify the discharge disposition (home, inpatient rehabilitation, skilled nursing inpatient facility [SNIF], and death) of patients hospitalized with COVID-19 based on hospital discharge summary notes. METHODS: Text mining and feature engineering were applied to unstructured text from hospital discharge summaries. The study included patients with COVID-19 discharged from 2 hospitals in the Boston, Massachusetts area (Massachusetts General Hospital and Brigham and Women's Hospital) between March 10, 2020, and June 30, 2020. The data were divided into a training set (70%) and hold-out test set (30%). Discharge summaries were represented as bags-of-words consisting of single words (unigrams), bigrams, and trigrams. The number of features was reduced during training by excluding n-grams that occurred in fewer than 10% of discharge summaries, and further reduced using least absolute shrinkage and selection operator (LASSO) regularization while training a multiclass logistic regression model. Model performance was evaluated using the hold-out test set. RESULTS: The study cohort included 1737 adult patients (median age 61 [SD 18] years; 55% men; 45% White and 16% Black; 14% nonsurvivors and 61% discharged home). The model selected 179 from a vocabulary of 1056 engineered features, consisting of combinations of unigrams, bigrams, and trigrams. The top features contributing most to the classification by the model (for each outcome) were the following: "appointments specialty," "home health," and "home care" (home); "intubate" and "ARDS" (inpatient rehabilitation); "service" (SNIF); "brief assessment" and "covid" (death). The model achieved a micro-average area under the receiver operating characteristic curve value of 0.98 (95% CI 0.97-0.98) and average precision of 0.81 (95% CI 0.75-0.84) in the testing set for prediction of discharge disposition. CONCLUSIONS: A supervised learning-based NLP approach is able to classify the discharge disposition of patients hospitalized with COVID-19. This approach has the potential to accelerate and increase the scale of research on patients' discharge disposition that is possible with EHR data.

Reasons for discontinuing insulin and factors associated with insulin discontinuation in patients with type 2 diabetes mellitus: a real-world evidence study.

BACKGROUND: Evidence suggests that insulin therapy of patients with type 2 diabetes mellitus (T2DM) is frequently discontinued. However, the reasons for discontinuing insulin and factors associated with insulin discontinuation in this patient population are not well understood. METHODS: We conducted a retrospective cohort study of adults with T2DM prescribed insulin between 2010 and 2017 at Partners HealthCare. Reasons for discontinuing insulin and factors associated with insulin discontinuation were studied using electronic medical records (EMR) data. Natural language processing (NLP) was applied to identify reasons from unstructured clinical notes. Factors associated with insulin discontinuation were extracted from structured EMR data and evaluated using multivariable logistic regression. RESULTS: Among 7009 study patients, 2957 (42.2%) discontinued insulin within 12 months after study entry. Most patients who discontinued insulin (2121 / 71.7%) had reasons for discontinuation documented. The most common reasons were improving blood glucose control (33.2%), achieved weight loss (18.5%) and initiation of non-insulin diabetes medications (16.7%). In multivariable analysis adjusted for demographics and comorbidities, patients were more likely to discontinue either basal or bolus insulin if they were on a basal-bolus regimen (OR 1.6, 95% CI 1.3 to 1.8; p <  0.001) or were being seen by an endocrinologist (OR 2.6; 95% CI 2.2 to 3.0; p <  0.001). CONCLUSIONS: In this large real-world evidence study conducted in an area with a high penetration of health insurance, insulin discontinuation countenanced by healthcare providers was common. In most cases it was linked to achievement of glycemic control, achieved weight loss and initiation of other diabetes medications. Factors associated with and stated reasons for insulin discontinuation were different from those previously described for non-adherence to insulin therapy, identifying it as a distinct clinical phenomenon.

CoVA: An Acuity Score for Outpatient Screening that Predicts Coronavirus Disease 2019 Prognosis.

BACKGROUND: We sought to develop an automatable score to predict hospitalization, critical illness, or death for patients at risk for coronavirus disease 2019 (COVID-19) presenting for urgent care. METHODS: We developed the COVID-19 Acuity Score (CoVA) based on a single-center study of adult outpatients seen in respiratory illness clinics or the emergency department. Data were extracted from the Partners Enterprise Data Warehouse, and split into development (n = 9381, 7 March-2 May) and prospective (n = 2205, 3-14 May) cohorts. Outcomes were hospitalization, critical illness (intensive care unit or ventilation), or death within 7 days. Calibration was assessed using the expected-to-observed event ratio (E/O). Discrimination was assessed by area under the receiver operating curve (AUC). RESULTS: In the prospective cohort, 26.1%, 6.3%, and 0.5% of patients experienced hospitalization, critical illness, or death, respectively. CoVA showed excellent performance in prospective validation for hospitalization (expected-to-observed ratio [E/O]: 1.01; AUC: 0.76), for critical illness (E/O: 1.03; AUC: 0.79), and for death (E/O: 1.63; AUC: 0.93). Among 30 predictors, the top 5 were age, diastolic blood pressure, blood oxygen saturation, COVID-19 testing status, and respiratory rate. CONCLUSIONS: CoVA is a prospectively validated automatable score for the outpatient setting to predict adverse events related to COVID-19 infection.