Leveraging Artificial Intelligence in the Peer Review of Spine Research Articles

Introduction: In the rapidly evolving field of neurosurgery, traditional peer review can delay the dissemination of critical research. This study assesses artificial intelligence’s (AI) ability to predict the acceptance or rejection of neurosurgical manuscripts, exploring its potential to enhance the peer review process.

Methods: Spine-related articles from Preprint.org and medRxiv.org were analyzed, comparing those later published with those not published after 12 months, presumed rejected. Each article was uploaded to ChatGPT 4o, Gemini, or Copilot with the prompt: “Based on the literature up to the date this article was posted, will it be accepted or rejected for publication following peer review? Please provide a yes or no answer.” Journal impact factors and cite scores were collected for accepted preprints. T-tests compared journal metrics between AI-predicted outcomes, and Chi-square analysis assessed AI accuracy. Additional analyses compared AI model performance

Results: A total of 20 preprints, 10 published and 10 presumed to be rejected, were included in the analysis. The average impact factor and cite score of accepted preprints were 3.48 ± 1.08 and 4.83 ± 1.37, respectively. The impact factor and cite score of journals corresponding to accepted preprints that were also accepted by ChatGPT and Gemini were not significantly different than those rejected by each platform (p=0.932 and p=0.490, and p=0.915 and p=0.376, respectively). Copilot accepted all preprints. ChatGPT, Gemini, and Copilot had significantly low overall performance (45%, 40%, and 50%, respectively, all p< 0.001) and variable performance in correctly accepting published preprints (90%, 50%, and 100%, respectively) or rejecting presumed to be rejected preprints (0%, 30%, and 0%, respectively).

Conclusion : Although AI shows moderate accuracy in predicting peer review outcomes, further algorithm development could streamline the peer review process. AI may also assist authors in journals selection, balancing high impact and acceptance potential for maximal chances of successful publication and appropriate contribution to the literature.