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.

Early-life exposure to Saharan dust storms and adolescence functional disability: Evidence from Cameroon.

The direct link between early-life dust storm exposure and later-in-life outcomes is not fully understood. This study examines the association of functional disability in a large sample of adolescent Cameroonians (N = 112,855) with in-utero and early childhood exposure to Saharan dust storms. Adjusting all estimations for temperature, precipitation, time and location fixed-effects, and person and family sociodemographic characteristics, we documented adverse effects on functional disability in female adolescents due to exposure to dense dust storms during the third gestation trimester and the second postnatal trimester. We also found suggestive evidence that an effect exists for the first as well as the third through fifth postnatal trimesters. In the third trimester of gestation and the second postnatal trimester, exposure to an average length dust storm with PM10 levels beyond 190 µg/m3 increased the likelihood of disability among female adolescents by approximately 229 (95 % CI: 10-464) in 100,000. The size of the adverse effects for the other periods followed similar patterns. These results show the value of creating infrastructures to mitigate or adapt to the effects of dust storms. These endeavors should focus on communities and populations in and around the Sahara where international organizations can play a role. In addition, establishing health data infrastructures not only improves public health but also advances our understanding of the long-term effects of dust storms. This study demonstrates the importance of research on the long-term effects of early-life exposure to dust storms and the need for additional work on this topic.

A replicable, open-source, data integration method to support national practice-based research & quality improvement systems.

OBJECTIVES: The Epilepsy Learning Healthcare System (ELHS) was created in 2018 to address measurable improvements in outcomes for people with epilepsy. However, fragmentation of data systems has been a major barrier for reporting and participation. In this study, we aimed to test the feasibility of an open-source Data Integration (DI) method that connects real-life clinical data to national research and quality improvement (QI) systems. METHODS: The ELHS case report forms were programmed as EPIC SmartPhrases at Mass General Brigham (MGB) in December 2018 and subsequently as EPIC SmartForms in June 2021 to collect actionable, standardized, structured epilepsy data in the electronic health record (EHR) for subsequent pull into the external national registry of the ELHS. Following the QI methodology in the Chronic Care Model, 39 providers, epileptologists and neurologists, incorporated the ELHS SmartPhrase into their clinical workflow, focusing on collecting diagnosis of epilepsy, seizure type according to the International League Against Epilepsy, seizure frequency, date of last seizure, medication adherence and side effects. The collected data was stored in the Enterprise Data Warehouse (EDW) without integration with external systems. We developed and validated a DI method that extracted the data from EDW using structured query language and later preprocessed using text mining. We used the ELHS data dictionary to match fields in the preprocessed notes to obtain the final structured dataset with seizure control information. For illustration, we described the data curated from the care period of 12/2018-12/2021. RESULTS: The cohort comprised a total of 1806 patients with a mean age of 43 years old (SD: 17.0), where 57% were female, 80% were white, and 84% were non-Hispanic/Latino. Using our DI method, we automated the data mining, preprocessing, and exporting of the structured dataset into a local database, to be weekly accessible to clinicians and quality improvers. During the period of SmartPhrase implementation, there were 5168 clinic visits logged by providers documenting each patient's seizure type and frequency. During this period, providers documented 59% patients having focal seizures, 35% having generalized seizures and 6% patients having another type. Of the cohort, 45% patients had private insurance. The resulting structured dataset was bulk uploaded via web interface into the external national registry of the ELHS. CONCLUSIONS: Structured data can be feasibly extracted from text notes of epilepsy patients for weekly reporting to a national learning healthcare system.

Noninvasive Detection of Hippocampal Epileptiform Activity on Scalp Electroencephalogram.

Importance: The hippocampus is a highly epileptogenic brain region, yet over 90% of hippocampal epileptiform activity (HEA) cannot be identified on scalp electroencephalogram (EEG) by human experts. Currently, detection of HEA requires intracranial electrodes, which limits our understanding of the role of HEA in brain diseases. Objective: To develop and validate a machine learning algorithm that accurately detects HEA from a standard scalp EEG, without the need for intracranial electrodes. Design, Setting, and Participants: In this diagnostic study, conducted from 2008 to 2021, EEG data were used from patients with temporal lobe epilepsy (TLE) and healthy controls (HCs) to train and validate a deep neural network, HEAnet, to detect HEA on scalp EEG. Participants were evaluated at tertiary-level epilepsy centers at 2 academic hospitals: Massachusetts General Hospital (MGH) or Brigham and Women's Hospital (BWH). Included in the study were patients aged 12 to 78 years with a clinical diagnosis of TLE and HCs without epilepsy. Patients with TLE and HCs with a history of intracranial surgery were excluded from the study. Exposures: Simultaneous intracranial EEG and/or scalp EEG. Main Outcomes and Measures: Performance was assessed using cross-validated areas under the receiver operating characteristic curve (AUC ROC) and precision-recall curve (AUC PR) and additional clinically relevant metrics. Results: HEAnet was trained and validated using data sets that were derived from a convenience sample of 141 eligible participants (97 with TLE and 44 HCs without epilepsy) whose retrospective EEG data were readily available. Data set 1 included the simultaneous scalp EEG and intracranial electrode recordings of 51 patients with TLE (mean [SD] age, 40.7 [15.9] years; 30 men [59%]) at MGH. An automatically generated training data set with 972 095 positive HEA examples was created, in addition to a held-out expert-annotated testing data set with 22 762 positive HEA examples. HEAnet's performance was validated on 2 independent scalp EEG data sets: (1) data set 2 (at MGH; 24 patients with TLE and 20 HCs; mean [SD] age, 42.3 [16.2] years; 17 men [39%]) and (2) data set 3 (at BWH; 22 patients with TLE and 24 HCs; mean [SD] age, 43.0 [14.4] years; 20 men [43%]). For single-event detection of HEA on data set 1, HEAnet achieved a mean (SD) AUC ROC of 0.89 (0.01) and a mean (SD) AUC PR of 0.39 (0.03). On external validation with data sets 2 and 3, HEAnet accurately distinguished TLE from HC (AUC ROC of 0.88 and 0.95, respectively) and predicted epilepsy lateralization with 100% and 92% accuracy, respectively. HEAnet tracked dynamic changes in HEA in response to seizure medication adjustments and performed comparably with human experts in diagnosing TLE from 1-hour scalp EEG recordings, diagnosing TLE in several individuals that experts missed. Without reducing specificity, addition of HEAnet to human expert EEG review increased sensitivity for diagnosing TLE in humans from 50% to 58% to 63% to 67%. Conclusions and Relevance: Results of this diagnostic study suggest that HEAnet provides a novel, noninvasive, quantitative, and clinically relevant biomarker of hippocampal hyperexcitability in humans.

Pediatric and adult neurologist perspectives on the challenges of sustaining a transfer clinic.

We present a novel epilepsy fellow-driven transfer clinic model and discuss the challenges experienced in finding sustainability; this is timely as many pioneering transition clinics are dissolving across North America. The goal of this clinic was to improve patient care and satisfaction, as measured by a post-visit telephone survey. Unfortunately, our transfer clinic model proved unsustainable due to several factors, broadly categorized as (1) cultural-societal differences between the pediatric and adult health care environments, (2) staffing issues, (3) lack of an established standardized process for transfer of care, and (4) financial and administrative barriers. We suggest potential solutions to these challenges, but the fate of transition and transfer of care clinics may ultimately depend on implementation of practice, policy, and/or financial guidelines.

Association of epileptiform abnormalities and seizures in Alzheimer disease.

OBJECTIVE: To examine the relationship between scalp EEG biomarkers of hyperexcitability in Alzheimer disease (AD) and to determine how these electric biomarkers relate to the clinical expression of seizures in AD. METHODS: In this cross-sectional study, we performed 24-hour ambulatory scalp EEGs on 43 cognitively normal elderly healthy controls (HC), 41 participants with early-stage AD with no history or risk factors for epilepsy (AD-NoEp), and 15 participants with early-stage AD with late-onset epilepsy related to AD (AD-Ep). Two epileptologists blinded to diagnosis visually reviewed all EEGs and annotated all potential epileptiform abnormalities. A panel of 9 epileptologists blinded to diagnosis was then surveyed to generate a consensus interpretation of epileptiform abnormalities in each EEG. RESULTS: Epileptiform abnormalities were seen in 53% of AD-Ep, 22% of AD-NoEp, and 4.7% of HC. Specific features of epileptiform discharges, including high frequency, robust morphology, right temporal location, and occurrence during wakefulness and REM, were associated with clinical seizures in AD. Multiple EEG biomarkers concordantly demonstrated a pattern of left temporal lobe hyperexcitability in early stages of AD, whereas clinical seizures in AD were often associated with bitemporal hyperexcitability. Frequent small sharp spikes were specifically associated with epileptiform EEGs and thus identified as a potential biomarker of hyperexcitability in AD. CONCLUSION: Epileptiform abnormalities are common in AD but not all equivalent. Specific features of epileptiform discharges are associated with clinical seizures in AD. Given the difficulty recognizing clinical seizures in AD, these EEG features could provide guidance on which patients with AD are at high risk for clinical seizures.

Racial and Ethnic Disparities in Postcardiac Arrest Targeted Temperature Management Outcomes.

OBJECTIVES: To evaluate racial and ethnic disparities in postcardiac arrest outcomes in patients undergoing targeted temperature management. DESIGN: Retrospective study. SETTING: ICUs in a single tertiary care hospital. PATIENTS: Three-hundred sixty-seven patients undergoing postcardiac arrest targeted temperature management, including continuous electroencephalogram monitoring. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Clinical variables examined in our clinical cohort included race/ethnicity, age, time to return of spontaneous circulation, cardiac rhythm at time of arrest, insurance status, Charlson Comorbidity Index, and time to withdrawal of life-sustaining therapy. CT at admission and continuous electroencephalogram monitoring during the first 24 hours were used as markers of early injury. Outcome was assessed as good (Cerebral Performance Category 1-2) versus poor (Cerebral Performance Category 3-5) at hospital discharge. White non-Hispanic ("White") patients were more likely to have good outcomes than white Hispanic/nonwhite ("Non-white") patients (34.4 vs 21.7%; p = 0.015). In a multivariate model that included age, time to return of spontaneous circulation, initial rhythm, combined electroencephalogram/CT findings, Charlson Comorbidity Index, and insurance status, race/ethnicity was still independently associated with poor outcome (odds ratio, 3.32; p = 0.003). Comorbidities were lower in white patients but did not fully explain outcomes differences. Nonwhite patients were more likely to exhibit signs of early severe anoxic changes on CT or electroencephalogram, higher creatinine levels and receive dialysis, but had longer duration to withdrawal of lifesustaining therapy. There was no significant difference in catheterizations or MRI scans. Subgroup analysis performed with patients without early electroencephalogram or CT changes still revealed better outcome in white patients. CONCLUSIONS: Racial/ethnic disparity in outcome persists despite a strictly protocoled targeted temperature management. Nonwhite patients are more likely to arrive with more severe anoxic brain injury, but this does not account for all the disparity.

Detection of mesial temporal lobe epileptiform discharges on intracranial electrodes using deep learning.

OBJECTIVE: Develop a high-performing algorithm to detect mesial temporal lobe (mTL) epileptiform discharges on intracranial electrode recordings. METHODS: An epileptologist annotated 13,959 epileptiform discharges from a dataset of intracranial EEG recordings from 46 epilepsy patients. Using this dataset, we trained a convolutional neural network (CNN) to recognize mTL epileptiform discharges from a single intracranial bipolar channel. The CNN outputs from multiple bipolar channel inputs were averaged to generate the final detector output. Algorithm performance was estimated using a nested 5-fold cross-validation. RESULTS: On the receiver-operating characteristic curve, our algorithm achieved an area under the curve (AUC) of 0.996 and a partial AUC (for specificity > 0.9) of 0.981. AUC on a precision-recall curve was 0.807. A sensitivity of 84% was attained at a false positive rate of 1 per minute. 35.9% of the false positive detections corresponded to epileptiform discharges that were missed during expert annotation. CONCLUSIONS: Using deep learning, we developed a high-performing, patient non-specific algorithm for detection of mTL epileptiform discharges on intracranial electrodes. SIGNIFICANCE: Our algorithm has many potential applications for understanding the impact of mTL epileptiform discharges in epilepsy and on cognition, and for developing therapies to specifically reduce mTL epileptiform activity.

Non-invasive Cognitive Enhancement in Epilepsy.

Epilepsy patients frequently experience cognitive difficulties, particularly in the domains of memory, attention, and executive function. Despite the frequency of these difficulties among epilepsy patients, current strategies to treat cognitive dysfunction are limited. We performed a systematic review of controlled trials of non-invasive cognitive enhancement in epilepsy. We identified studies examining the efficacy of pharmacological agents, namely the acetylcholinesterase inhibitors donepezil and galantamine, the NMDA non-competitive antagonist memantine, and the stimulant methylphenidate, as well as non-invasive non-pharmacological transcranial magnetic stimulation (TMS) and transcranial direct current stimulation (tDCS). We highlight the data currently available and the limitations of the current literature.

High-throughput brain activity mapping and machine learning as a foundation for systems neuropharmacology.

Technologies for mapping the spatial and temporal patterns of neural activity have advanced our understanding of brain function in both health and disease. An important application of these technologies is the discovery of next-generation neurotherapeutics for neurological and psychiatric disorders. Here, we describe an in vivo drug screening strategy that combines high-throughput technology to generate large-scale brain activity maps (BAMs) with machine learning for predictive analysis. This platform enables evaluation of compounds' mechanisms of action and potential therapeutic uses based on information-rich BAMs derived from drug-treated zebrafish larvae. From a screen of clinically used drugs, we found intrinsically coherent drug clusters that are associated with known therapeutic categories. Using BAM-based clusters as a functional classifier, we identify anti-seizure-like drug leads from non-clinical compounds and validate their therapeutic effects in the pentylenetetrazole zebrafish seizure model. Collectively, this study provides a framework to advance the field of systems neuropharmacology.

Fungal infections of the central nervous system.

Fungal infections of the central nervous system have manifold presentations and courses that depend largely on both host and organism characteristics. Although subjects with impaired immunity are generally at higher risk for severe disease, several fungal organisms are considered primary pathogens and can also cause disease in otherwise immunocompetent individuals. Herein, we describe the epidemiology, presentation, diagnosis, and management of central nervous system complications of several fungal pathogens.

Hemorrhagic and ischemic stroke secondary to herpes simplex virus type 2 meningitis and vasculopathy.

Herpes simplex virus type 2 (HSV-2) meningitis dogmatically is benign and self-limited in the immune competent patient. However, we describe how left untreated HSV-2 meningitis can be complicated by vasculitis and both ischemic and hemorrhagic stroke. We report a 57-year-old woman with lymphocytic meningitis complicated by ischemic stroke and intracerebral hemorrhage in the setting of vasculopathy and HSV-2 DNA detected in CSF successfully treated with acyclovir and corticosteroids. Subsequent angiographic magnetic resonance imaging revealed improvement in the vasculopathy after treatment. This case demonstrates that HSV-2 meningitis may take a less benign course and further provides the first evidence of angiographic improvement in addition to clinical improvement after definitive treatment.

Modifications at the C-terminus to improve pyrrole-imidazole polyamide activity in cell culture.

Pyrrole-imidazole (Py-Im) hairpin polyamides are a class of small molecule DNA minor groove binding compounds that have been shown to modulate endogenous gene expression in cell culture. Gene regulation by polyamides requires efficient cellular uptake and nuclear localization properties for candidate compounds. To further optimize Py-Im polyamides for enhanced potency in cell culture, a focused library of polyamides possessing various modifications at the C-terminus was synthesized and tested. Comparison of polyamide biological activity in two cell lines revealed tolerance for structural modifications and agreement in activity trends between cell lines. The use of an oxime linkage between the polyamide and an aromatic functionality on the C-terminus resulted in a approximately 20-fold increase in the potency of polyamides targeted to the androgen response element (ARE) in LNCaP cells by measuring AR-activated PSA expression.

Modulating hypoxia-inducible transcription by disrupting the HIF-1-DNA interface.

Transcription mediated by hypoxia-inducible factor (HIF-1) contributes to tumor angiogenesis and metastasis but is also involved in activation of cell-death pathways and normal physiological processes. Given the complexity of HIF-1 signaling, it could be advantageous to target a subset of HIF-1 effectors rather than the entire pathway. We compare the genome-wide effects of three molecules that each interfere with the HIF-1-DNA interaction: a polyamide targeted to the hypoxia response element, small interfering RNA targeted to HIF-1alpha, and echinomycin, a DNA-binding natural product with a similar but less specific sequence preference than the polyamide. The polyamide affects a subset of hypoxia-induced genes consistent with its binding site preferences. For comparison, HIF-1alpha siRNA and echinomycin each affect the expression of nearly every gene induced by hypoxia. Remarkably, the total number of genes affected by either polyamide or HIF-1alpha siRNA over a range of thresholds is comparable. The data show that polyamides can be used to affect a subset of a pathway regulated by a transcription factor. In addition, this study offers a unique comparison of three complementary approaches towards exogenous control of endogenous gene expression.

Improved nuclear localization of DNA-binding polyamides.

Regulation of endogenous genes by DNA-binding polyamides requires effective nuclear localization. Previous work employing confocal microscopy to study uptake of fluorophore-labeled polyamides has demonstrated the difficulty of predicting a priori the nuclear uptake of a given polyamide. The data suggest that dye identity influences uptake sufficiently such that a dye-conjugate cannot be used as a proxy for unlabeled analogs. Polyamides capable of nuclear localization unaided by fluorescent dyes are desirable due to size and other limitations of fluorophores. Recently, a polyamide-fluorescein conjugate targeted to the hypoxia response element (HRE) was found to inhibit vascular endothelial growth factor (VEGF) expression in cultured HeLa cells. The current study uses inhibition of VEGF expression as a biological read-out for effective nuclear localization of HRE-targeted polyamides. We synthesized a focused library of non-fluorescent, HRE-targeted polyamides in which the C-terminus 'tail' has been systematically varied. Members of this library bind the HRE with affinities comparable or superior to that of the fluorescein-labeled analog. Although most library members demonstrate modest or no biological activity, two non-fluorescent polyamides are reported with activity rivaling that of the previously reported fluorescein-labeled polyamide. We also show evidence that promoter occupancy by HIF-1, the transcription factor that binds the HRE, is inhibited by HRE-targeted polyamides.