Click to open pdf in new window
Dr. Maximilian Maywald
Department of Psychiatry and Psychotherapy, LMU University Hospital
Systematic reviews are crucial for synthesizing existing evidence on a given topic. They help identify research gaps in the literature and guide future research. However, as the volume and complexity of published studies continue to rise, traditional approaches to systematic reviews have become increasingly time-consuming and less efficient.
AI-based tools present an opportunity to streamline the review process and enhance efficiency.
This study compares the quality of outcomes and time savings during title and abstract screening using elicit.ai software versus conventional screening methods. Identical sets of articles were screened using both approaches. Disagreements were resolved through consensus with a third reviewer.
The study also discusses the advantages and potential risks, including the possibility of missing relevant studies, when employing AI-assisted tools.