Efficient evidence-based selection for clinical researchers in Traditional Chinese Medicine

Abstract Purpose: Searching and selecting clinical evidence for systematic reviews (SRs) or clinical guidelines are vital for researchers in traditional Chinese medicine (TCM). The process can be time-consuming and resource-intensive, requiring significant effort from researchers. In this study, we introduce a new method for evidence-based selection that combines artificial intelligence (AI) and human efforts to achieve both swift and precise for TCM practitioners. Methods: We use the knowledge engineer (KE) approach and a series of Boolean logic codes to select potential evidence automatically and accurately, with minimal human intervention. The selection details are recorded in real-time, enabling researchers to backtrack and verify the accuracy of the selection process. We apply the new approach in ten high-quality systematic reviews of randomly selected with TCM topics in the Chinese language. To evaluate the method's effectiveness, we compare the screen time and accuracy of the traditional selection method with the new process. Results: The results show that the new method can accurately select potential literature under the same criteria while taking less time. Moreover, the new approach can identify more related evidence literature in some cases while tracking the selection progress for future reference. This study also identifies traditional screening methods are subjective and may lead to error inclusion of literature that does not meet the standards. The new method provides a more accurate and efficient way to select potential clinical evidence for TCM practitioners, outperforming traditional methods that rely solely on human effort. Conclusion: We offer a novel approach to select clinical evidence for reviews and guidelines in TCM that can significantly reduce the workload for researchers. Our method holds promise for improving the efficiency and accuracy of evidence-based selection and may be used by editors to check the quality of manuscripts in the future.