From radiology findings to artificial intelligence-powered impressions: A retrospective study on the comparative performance of recent large language models
Bibliographic record
Abstract
Large language models (LLMs), a revolutionary breakthrough in Artificial Intelligence, can be leveraged to automatically generate the impressions for radiology reports, which usually requires time, effort, and training. Our objective was to evaluate the performance of five recent LLMs (GPT-4, GPT-4o mini, Gemini 1.5 – Pro, Gemini 1.5 – Flash, and Llama 3.1) for impression generation. In this retrospective study, 100 radiology reports were sampled (20 from each of the report-groups 0-400, 400-800, 800-1,200, 1,200-2,000, and 2,000–8,000 based on character count of findings section) from the publicly available “BioNLP 2023 report summarization” dataset (collected between 2001-2016, training subset of size 59,320 considered for sampling), sourced from PhysioNet. Then each of the five LLMs was zero-shot prompted to generate impressions using the findings from the sample. Generated impressions were evaluated: (a) subjectively for coherence, comprehensiveness, conciseness, and medical harmfulness by two radiology fellows and a large reasoning model (LRM) Gemini 2.5 – Pro, and (b) objectively using a composite accuracy metric (ROUGE-1, BLEU, and Cosine Similarity) against the original human expert-generated impressions. The LLMs were ranked according to the percentage agreement ranking of subjective scores and composite scores. Statistical tests (Friedman test and post-hoc Nemenyi test) were used to assess inter-model differences. The top-ranked models were Gemini 1.5 – Pro, GPT-4, and Gemini 1.5 – Flash. Performance varied across models for both human and LRM raters (Friedman test: Human P < 1.82 × 10⁻⁶; LRM P < 9.10 × 10⁻⁴⁰). Composite accuracy scores were significantly higher for the top three models (0.69, 0.68, 0.68) versus others (0.65; Nemenyi P < 1.11 × 10⁻¹⁶). The LRM aligned closely with human raters (2.15% complete disagreement) and identified all human-rated inaccurate impressions. Gemini 1.5 – Pro outperformed GPT-4, in terms of coherence, comprehensiveness, and medical harmfulness, at lower cost. Human and LRM evaluations were generally consistent, though the LRM was more conservative.
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 imitationNot 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.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.015 | 0.084 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.002 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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.
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 source (direct Gemma or distilled Codex), 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".