MétaCan
Menu
Back to cohort
Record W4393157213 · doi:10.1609/aaai.v38i20.30205

MedAlign: A Clinician-Generated Dataset for Instruction Following with Electronic Medical Records

2024· article· en· W4393157213 on OpenAlexaff
Scott L. Fleming, Alejandro Lozano, William J. Haberkorn, Jenelle Jindal, Eduardo Pontes Reis, Rahul Thapa, Louis Blankemeier, Julian Z. Genkins, Ethan Steinberg, Ashwin Nayak, Birju Patel, Chia‐Chun Chiang, Alison Callahan, Zepeng Huo, Sergios Gatidis, Scott Adams, Oluseyi Fayanju, Shreya Shah, Thomas Savage, Ethan Goh, Akshay Chaudhari, Nima Aghaeepour, Christopher Sharp, Michael A. Pfeffer, Percy Liang, Jonathan H. Chen, Keith Morse, Emma Brunskill, Jason Fries, Nigam H. Shah

Bibliographic record

VenueProceedings of the AAAI Conference on Artificial Intelligence · 2024
Typearticle
Languageen
FieldHealth Professions
TopicElectronic Health Records Systems
Canadian institutionsArtificial Intelligence in Medicine (Canada)
FundersNational Institute of Biomedical Imaging and BioengineeringNational Institute of Arthritis and Musculoskeletal and Skin DiseasesNational Institute of General Medical SciencesNational Heart, Lung, and Blood Institute
KeywordsMedical recordElectronic medical recordComputer scienceMedical educationPsychologyData scienceMedicineInformation retrievalInternet privacyInternal medicine

Abstract

fetched live from OpenAlex

The ability of large language models (LLMs) to follow natural language instructions with human-level fluency suggests many opportunities in healthcare to reduce administrative burden and improve quality of care. However, evaluating LLMs on realistic text generation tasks for healthcare remains challenging. Existing question answering datasets for electronic health record (EHR) data fail to capture the complexity of information needs and documentation burdens experienced by clinicians. To address these challenges, we introduce MedAlign, a benchmark dataset of 983 natural language instructions for EHR data. MedAlign is curated by 15 clinicians (7 specialities), includes clinician-written reference responses for 303 instructions, and provides 276 longitudinal EHRs for grounding instruction-response pairs. We used MedAlign to evaluate 6 general domain LLMs, having clinicians rank the accuracy and quality of each LLM response. We found high error rates, ranging from 35% (GPT-4) to 68% (MPT-7B-Instruct), and 8.3% drop in accuracy moving from 32k to 2k context lengths for GPT-4. Finally, we report correlations between clinician rankings and automated natural language generation metrics as a way to rank LLMs without human review. MedAlign is provided under a research data use agreement to enable LLM evaluations on tasks aligned with clinician needs and preferences.

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 imitation

Not 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.

metaresearch head score (Codex)0.004
metaresearch head score (Gemma)0.002
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: Theoretical or conceptual
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.477
Threshold uncertainty score0.873

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0040.002
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0010.000
Bibliometrics0.0000.001
Science and technology studies0.0010.000
Scholarly communication0.0000.000
Open science0.0010.000
Research integrity0.0000.002
Insufficient payload (model declined to judge)0.0000.000

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.

Opus teacher head0.151
GPT teacher head0.461
Teacher spread0.311 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designTheoretical or conceptual
Domainnot available
GenreEmpirical

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".

Quick stats

Citations46
Published2024
Admission routes1
Has abstractyes

Explore more

Same venueProceedings of the AAAI Conference on Artificial IntelligenceSame topicElectronic Health Records SystemsFrench-language works237,207