MétaCan
Menu
← Back to cohort
Record W4413846962 · doi:10.2196/67661

Rapidly Benchmarking Large Language Models for Diagnosing Comorbid Patients: Comparative Study Leveraging the LLM-as-a-Judge Method

2025· article· en· W4413846962 on OpenAlexvenueno aff
Peter Sarvari, Zaid Al-Fagih

Bibliographic record

VenueJMIRx Med · 2025
Typearticle
Languageen
FieldMedicine
TopicArtificial Intelligence in Healthcare and Education
Canadian institutionsnot available
Fundersnot available
KeywordsBenchmarkingComputer scienceNatural language processingData scienceArtificial intelligencePsychologyBusiness

Abstract

fetched live from OpenAlex

Background: On average, 1 in 10 patients die because of a diagnostic error, and medical errors represent the third largest cause of death in the United States. While large language models (LLMs) have been proposed to aid doctors in diagnoses, no research results have been published comparing the diagnostic abilities of many popular LLMs on a large, openly accessible real-patient cohort. Objective: In this study, we set out to compare the diagnostic ability of 18 LLMs from Google, OpenAI, Meta, Mistral, Cohere, and Anthropic, using 3 prompts, 2 temperature settings, and 1000 randomly selected Medical Information Mart for Intensive Care-IV (MIMIC-IV) hospital admissions. We also explore improving the diagnostic hit rate of GPT-4o 05-13 with retrieval-augmented generation (RAG) by utilizing reference ranges provided by the American Board of Internal Medicine. Methods: We evaluated the diagnostic ability of 21 LLMs, using an LLM-as-a-judge approach (an automated, LLM-based evaluation) on MIMIC-IV patient records, which contain final diagnostic codes. For each case, a separate assessor LLM ("judge") compared the predictor LLM's diagnostic output to the true diagnoses from the patient record. The assessor determined whether each true diagnosis was inferable from the available data and, if so, whether it was correctly predicted ("hit") or not ("miss"). Diagnoses not inferable from the patient record were excluded from the hit rate analysis. The reported hit rate was defined as the number of hits divided by the total number of hits and misses. The statistical significance of the differences in model performance was assessed using a pooled z-test for proportions. Results: Gemini 2.5 was the top performer with a hit rate of 97.4% (95% CI 97.0%-97.8%) as assessed by GPT-4.1, significantly outperforming GPT-4.1, Claude-4 Opus, and Claude Sonnet. However, GPT-4.1 ranked the highest in a separate set of experiments evaluated by GPT-4 Turbo, which tended to be less conservative than GPT-4.1 in its assessments. Significant variation in diagnostic hit rates was observed across different prompts, while changes in temperature generally had little effect. Finally, RAG significantly improved the hit rate of GPT-4o 05-13 by an average of 0.8% (P<.006). Conclusions: While the results are promising, more diverse datasets and hospital pilots, as well as close collaborations with physicians, are needed to obtain a better understanding of the diagnostic abilities of these models.

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 machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.014
metaresearch head score (Gemma)0.054
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.014
Threshold uncertainty score0.075

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0140.054
Meta-epidemiology (narrow)0.0020.001
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0030.001
Science and technology studies0.0010.001
Scholarly communication0.0020.002
Open science0.0030.002
Research integrity0.0020.002
Insufficient payload (model declined to judge)0.0010.001

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 head0.164
GPT teacher head0.500
Teacher spread0.336 · 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 source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
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

Citations3
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueJMIRx Med→Same topicArtificial Intelligence in Healthcare and Education→French-language works237,207→