ChatGPT-Exploring Its Role in Clinical Chemistry.
Bibliographic record
Abstract
OBJECTIVE: To evaluate the utility of artificial intelligence-powered language models (ChatGPT 3.5 and GPT-4) compared to trainees and clinical chemists in responding to common laboratory questions in the broad area of Clinical Chemistry. METHODS: 35 questions from real-life case scenarios, clinical consultations, and clinical chemistry testing questions were used to evaluate ChatGPT 3.5, and GPT-4 alongside clinical chemistry trainees (residents/fellows) and clinical chemistry faculty. The responses were scored based on category and based on years of experience. RESULTS: The Senior Chemistry Faculty demonstrated superior accuracy with 100% of correct responses compared to 90.5%, 82.9%, and 71.4% of correct responses from the junior chemistry faculty, fellows, and residents respectively. They all outperformed both ChatGPT 3.5 and GPT-4 which generated 60% and 71.4% correct responses respectively. Of the sub-categories examined, ChatGPT 3.5 achieved 100% accuracy in endocrinology while GPT-4 did not achieve 100% accuracy in any subcategory. GPT-4 was overall better than ChatGPT 3.5 by generating similar correct responses as residents (71.4%) but performed poorly to human participants when both partially correct and incorrect indices were considered. CONCLUSION: Despite all the advances in AI-powered language models, ChatGPT 3.5 and GPT-4 cannot replace a trained pathologist in answering clinical chemistry questions. Caution should be observed by people, especially those not trained in clinical chemistry, to interpret test results using chatbots.
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 | 0.062 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.003 | 0.003 |
| Open science | 0.002 | 0.003 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.007 | 0.004 |
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".