Objective: This article presents the Harvard Electroencephalography Database (HEEDB), a large‐scale, deidentified, and standardized electroencephalographic (EEG) resource supporting artificial intelligence‐driven and reproducible research in epilepsy and broader clinical neuroscience.Methods: HEEDB aggregates more than 280 000 EEG recordings from more than 108 000 patients across four Harvard‐affiliated hospitals. Data are harmonized using the Brain Imaging Data Structure and hosted on the Brain Data Science Platform. EEG data are linked with clinical notes, International Classification of Diseases, 10th Revision codes, medications, and EEG reports. Deidentification follows Health Insurance Portability and Accountability Act Safe Harbor standards.Results: The database includes routine, epilepsy monitoring unit, and intensive care unit EEGs across all age groups, with 73% linked to deidentified clinical reports and 96% of those matched to recordings. Findings are extracted using expert curation, regular expressions, and medical natural language processing models. Auxiliary data include diagnoses, medications, and hospital course, supporting multimodal analysis.Significance: HEEDB fills a critical gap in EEG data availability for epilepsy research. By enabling large‐scale, privacy‐compliant, and clinically relevant analysis, it accelerates the development of diagnostic tools, improves training datasets for machine learning, and promotes data‐sharing in alignment with FAIR (Findable, Accessible, Interoperable, Reusable) and National Institutes of Health data policies.

Patterns of electrical activity in the brain (EEG) during sleep are sensitive to various health conditions even at subclinical stages. The objective of this study was to estimate sleep EEG-predicted incidence of future neurologic, cardiovascular, psychiatric, and mortality outcomes. This is a retrospective cohort study with 2 data sets. The Massachusetts General Hospital (MGH) sleep data set is a clinic-based cohort, used for model development. The Sleep Heart Health Study (SHHS) is a community-based cohort, used as the external validation cohort. Exposure is good, average, or poor sleep defined by quartiles of sleep EEG-predicted risk. The outcomes include ischemic stroke, intracranial hemorrhage, mild cognitive impairment, dementia, atrial fibrillation, myocardial infarction, type 2 diabetes, hypertension, bipolar disorder, depression, and mortality. Diagnoses were based on diagnosis codes, brain imaging reports, medications, cognitive scores, and hospital records. We used the Cox survival model with death as the competing risk. There were 8673 participants from MGH and 5650 from SHHS. For all outcomes, the model-predicted 10-year risk was within the 95% confidence interval of the ground truth, indicating good prediction performance. When comparing participants with poor, average, and good sleep, except for atrial fibrillation, all other 10-year risk ratios were significant. The model-predicted 10-year risk ratio closely matched the observed event rate in the external validation cohort. The incidence of health outcomes can be predicted by brain activity during sleep. The findings strengthen the concept of sleep as an accessible biological window into unfavorable brain and general health outcomes.

Study Objectives: To develop and validate a Complete Artificial Intelligence Sleep Report system (CAISR), a system for comprehensive automated sleep analysis, including sleep staging, arousal detection, apnea identification, and limb movement analysis. Methods: We utilized a large diverse dataset from four cohorts (MGH, MESA, MrOS, SHHS) comprising 25,749 participants to develop CAISR. Following American Academy of Sleep Medicine (AASM) guidelines, CAISR performs four tasks: it stages sleep into five categories (Wake, NREM 1, NREM 2, NREM 3, REM), detects arousals, detects and classifies breathing events (Obstructive Apnea, Central Apnea, Mixed Apnea, Hypopnea, and RERA), and detects limb movements and categorizes them as periodic or isolated. We tested CAISR against multiple datasets independently annotated by multiple experts, including UPenn (69 subjects, 6 experts), BITS (98 subjects, 3 experts), and Stanford (100 subjects, three experts). Sleep staging and arousal detection were accomplished using customized deep neural networks, while breathing event detection and classification and limb movement analysis were accomplished using rule-based signal processing approaches. We quantified CAISR performance with three metrics: Cohen’s Kappa, area under the receiver operating characteristic curve (AUROC), and area under the precision-recall curve (AUPRC). To determine whether CAISR performed on par with human experts, we compared expert inter-rater reliability (IRR) with algorithm-expert IRR. Results: The CAISR model showed strong overall performance across the four tasks: sleep staging, arousal detection, apnea detection, and limb movement detection. In sleep staging, the model achieved AUROC values ranging from 0.82 to 0.97 and AUPRC values between 0.63 and 0.90 across the BITS, Stanford, and Penn datasets, indicating high classification accuracy. The Kappa agreement analysis showed that in the BITS and Stanford datasets, CAISR outperformed human experts, with non-overlapping confidence intervals indicating superiority (Kappa values around 0.7 to 0.8 for CAISR vs. experts). In the Penn dataset, the model’s performance was comparable to experts, with overlapping confidence intervals suggesting non-inferiority. For arousal detection, the model maintained reliable performance, with AUROC values ranging from 0.83 to 0.94 and AUPRC values from 0.67 to 0.85, and Kappa analysis showing overlapping confidence intervals, indicating comparable performance to experts in both the BITS and Stanford datasets (Kappa values for CAISR around 0.6 to 0.75). In apnea detection, including the detection of obstructive, central, and mixed apnea, the CAISR model achieved competitive results in the BITS dataset with AUROC values between 0.81 and 0.95 and AUPRC values between 0.58 and 0.82, but in the Stanford dataset, it underperformed compared to human experts, as shown by non-overlapping confidence intervals and lower Kappa values (around 0.55 to 0.65). Finally, in limb movement detection, the model demonstrated superior performance in the BITS dataset, with AUROC values of 0.9 to 0.96 and AUPRC values between 0.75 and 0.85, and Kappa analysis indicating significantly higher reliability compared to experts (CAISR Kappa around 0.8, with non-overlapping confidence intervals). In the Stanford dataset, CAISR’s performance was comparable to experts, with overlapping confidence intervals suggesting non-inferiority (Kappa values around 0.65 to 0.7). Overall, the CAISR model consistently exhibited high classification performance and reliability across tasks, often matching or surpassing expert-level performance, with particularly strong results in sleep staging and limb detection. Conclusions: The CAISR model demonstrated high classification accuracy and reliability across sleep staging, arousal, apnea, and limb movement detection tasks, matching or surpassing human expert performance. Human errors and systematic biases in the annotation of micro-events during sleep, such as arousal and apnea detection, likely contributed to variability in expert performance, while the CAISR model showed more consistent results, reducing the impact of these biases and increasing overall reliability across task.

Sleep underpins cognition, disease prevention, and overall brain health, yet objective, integrative biomarkers of brain health remain lacking. We hypothesized that overnight sleep electroencephalography (EEG) could provide a substrate for such a biomarker. We asked whether a newly developed, end-to-end, data-driven deep learning framework for sleep EEG can learn a latent representation of brain health and distill it into a single score relevant to cognition, disease status, and mortality. We analyzed 36,000 polysomnography recordings from 27,000 subjects from six cohorts. EEG data were represented as one-dimensional time series or a two-dimensional time-frequency spectrogram. A multitask deep neural network, trained end-to-end without expert-defined features, learned a 1024-dimensional brain health latent space and jointly predicted cognitive performance, disease status, and sleep metrics. The latent representation was additionally distilled into a single brain health score. We compared performance with demographic baselines, conventional EEG metrics (e.g., rapid eye movement fraction, spindle density), and classic multivariate machine learning approaches. The deep learning–derived brain health scores consistently surpassed demographic and expert-defined EEG feature models. For cognitive outcomes, correlations (r) rose from small (demographic-only) to moderate (up to r=0.40), while disease classification areas under the receiver operator curve improved from 0.50–0.55 at baseline to 0.65–0.75. In age-adjusted Cox models, a one-standard-deviation increase in the brain health score was associated with a 31%–35% reduced risk of mortality (hazard ratio 0.65 to 0.69; P<0.0001), topping conventional EEG metrics. Gains over classic machine learning, plus latent space visualization, indicated that both established physiological markers and novel EEG features drove enhanced performance. A multitask, end-to-end deep learning approach generated an interpretable, sleep-derived brain health biomarker. By modeling cognition, disease, and mortality, this framework provides a robust index of brain health and may be extended to additional modalities, further enhancing its clinical utility. (Funded by the National Institutes of Health and others.)

The Harvard-Emory ECG Database (HEEDB) is currently the largest open-access collection of 12-lead electrocardiogram (ECG) recordings, developed through a collaboration between Harvard and Emory University. The database consists of 10,608,417 ECG recordings from 1,818,247 patients from Massachusetts General Hospital (MGH) and 998,844 recordings from 349,548 patients from Emory University Hospital (EUH) collected between 1980 and 2022 in clinical settings as part of routine patient care. The ECGs are 10-second, 12-lead recordings sampled at either 250 or 500 Hz, and stored in WFDB format. Each ECG is linked to demographic metadata (age, sex, race, ethnicity, education), along with deidentified acquisition dates, last visit dates, and death dates, when available. The dataset includes three forms of Marquette^TM 12SL Analysis Software annotations: (1) batch-reprocessed diagnostic labels generated using the latest available 12SL software (version 24); (2) the original 12SL outputs from the time of ECG acquisition; and (3) the corresponding physician overreads. Additionally, the dataset includes associated ICD-9 and ICD-10 codes with the corresponding diagnosis dates. This database represents a large, diverse multi-center collection on which machine learning algorithms can be trained and tested for performance and bias.

Sleep is a fundamental biological process essential to health, particularly during early life when sleep patterns are developing and sleep disorders are common. Yet pediatric sleep research is hindered by a lack of large-scale, high-quality polysomnography (PSG) datasets. To address this need, we introduce the Boston Children’s Hospital (BCH) Sleep Corpus—the largest pediatric PSG dataset available—comprising 15 695 overnight recordings from 12 640 unique patients (median age ~ 6 years). The dataset includes 16.7 million annotated sleep stages, 2.25 million respiratory, arousal, and limb movement events, and over 11 000 patient diagnoses linked through de-identified electronic health records. Each PSG has a median duration of 8.9 hours, totaling 139 208 hours of EEG data. Sleep staging follows American Academy of Sleep Medicine guidelines and reveals age-related trends: REM sleep decreases from 33.5% in neonates to 16.3% in teenagers, while N2 sleep increases from 21.7% to 35.4%. Central apneas decline with age, while obstructive hypopneas and respiratory effort related arousals events rise. Limb movements are not scored in <1 yr but remain at around 30 per PSG across older age groups. We also present age- and region-specific EEG spectral norms and respiratory event trends across the pediatric age range. The dataset is organized in Brain Imaging Data Structure (BIDS) format and publicly available via the Brain Data Science Platform. The dataset provides a valuable resource for improving our scientific understanding of pediatric sleep and developing automated PSG analysis with artificial intelligence tools.