Content Validity of AI-Generated Medical Information on Idiopathic Pulmonary Fibrosis (IPF): A Comparative Analysis of ChatGPT-4 and Gemini 1.5 Pro
Bibliographic record
Abstract
Background: IPF is characterized by progressive declining respiratory function and quality of life, with high mortality. Large language models (LLMs) produce coherent medical information, but their accuracy, readability, and adherence to IPF guidelines remain unconfirmed. Aim: To evaluate the reliability and accuracy of LLMs in generating medically and clinically relevant content related to IPF. Methods: A comparative analysis of ChatGPT-4 and Gemini 1.5 Pro responses about IPF-related 23 questions from ATS/ERS/JRS/ALAT guidelines was conducted. Six independent ILD experts assessed responses for accuracy (DISCERN), reliability (JAMA Benchmark Criteria), readability (Flesch-Kincaid), and guidelines adherence. Mann-Whitney U tests and intraclass correlation coefficients (ICC) were used to compare model performance. Results: Both LLMs provided partially sufficient responses, with a median JAMA Benchmark score of 2 for both models (p = 0.24). Gemini 1.5 Pro generated higher-quality treatment-related responses compared to ChatGPT-4, as reflected by significantly higher DISCERN scores of 56 and 43, respectively (p < 0.001). Regarding readability, both models required college-level comprehension. The ICC analysis revealed significant inter-rater variability, with ChatGPT-4 demonstrating lower agreement (ICC = 0.361) than Gemini 1.5 Pro (ICC = 0.813). Conclusion: While both models offer coherent medical information, their reliability remains suboptimal. Further research should focus on improving AI readability on IPF for practical integration into clinical practice.
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 | 0.026 | 0.140 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.003 | 0.002 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.003 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".