NeuroDiscovery AI database: Comprehensive EHR dataset for Neurology
Bibliographic record
Abstract
Abstract Purpose The NeuroDiscovery AI database is a comprehensive real-world data (RWD) repository containing de-identified electronic health record (EHR) data from U.S.-based neurology outpatient clinics. The structured data encompasses sociodemographic details, clinical examinations, social, medical, and lifestyle histories, International Classification of Diseases (ICD-9/ICD-10) diagnoses, and prescribed medications. Additionally, the database integrates neuroimaging data and laboratory results, providing a robust resource for clinical research. This paper describes a subset of the NeuroDiscovery AI dataset and outlines the processes involved in its development. Participants As of October 15, 2024, the dataset includes EHR data from 355,791 patients, of whom 40.72% are male. Over 40.06% of the patients are aged 60 or older, spanning across 14,797 distinct diagnosis codes. The data represents more than 15 years of longitudinal patient information, with 26.87% of patients classified as active (defined as having had clinical encounters within the last 18 months). The median follow-up duration for active patients is 19.54 months. Limitation The large sample size, rigorous data processing, and robust data security of the NeuroDiscovery AI dataset are key strengths, enabling comprehensive studies on disease progression, treatment responses, and long-term outcomes in neurology. The dataset aligns closely with published demographic trends for various neurological conditions, including a female predominance in migraines, multiple sclerosis, and vertigo, with slight variations in age and gender distribution for conditions such as ALS. However, challenges remain, including missing data and data heterogeneity. Ongoing efforts to expand and diversify the dataset aim to improve its applicability and representativeness. Future plan The NeuroDiscovery AI dataset will expand by incorporating data from more providers and improving diversity, aiming to become one of the largest neurology-focused datasets. The platform will continue to evolve into a comprehensive analytical tool, integrating cohort building and data interrogation functionalities to streamline clinical workflows. These enhancements will enable faster, more accurate decision-making, and future efforts will focus on identifying key trends in neurological conditions and patient outcomes.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.004 | 0.008 |
| Research integrity | 0.000 | 0.002 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".