ChatGPT-4.0 as a Tool for Automated Review of Ethics and Transparency in Biomedical Literature
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.
View abstract on PubMed
Summary
This summary is machine-generated.ChatGPT-4.0 effectively identifies missing ethics statements in public health journals, showing high sensitivity. However, precision varies, necessitating human oversight for robust editorial checks and maintaining ethical standards.
Area Of Science
- Public Health
- Bibliometrics
- Artificial Intelligence in Publishing
Background
- Growing interest in AI, specifically large language models (LLMs), for streamlining editorial processes.
- Focus on ethical and transparency reporting in public health journals.
- Need to evaluate AI capabilities in manuscript assessment.
Purpose Of The Study
- Evaluate ChatGPT-4.0's accuracy in detecting missing ethical and transparency statements.
- Compare performance between high-ranked (Q1) and low-ranked (Q4) public health journals.
Main Methods
- Analysis of articles from Q1 and Q4 public health journals using ChatGPT-4.0.
- Assessment for essential ethical components: ethics approval, informed consent, animal ethics, conflicts of interest, funding, and data sharing.
- Calculation of performance metrics: sensitivity, recall, and precision.
Main Results
- ChatGPT-4.0 demonstrated high sensitivity and recall for all evaluated ethical components.
- Precision varied, with high accuracy for data availability (0.96) but lower for funding statements (0.16).
- Q4 journals showed a significant increase in missing statements, especially for open data sharing, ethics approval, and informed consent.
Conclusions
- ChatGPT-4.0 shows promise for preliminary screening of missing ethics statements with high accuracy.
- Precision limitations necessitate complementary human review.
- Recommends balanced integration of AI and human judgment to enhance editorial checks and uphold ethical standards.
These videos have been matched automatically. Contact us if they are not relevant.
These videos have been matched automatically. Contact us if they are not relevant.