Improving search strategies in bibliometric studies on machine learning in renal medicine.

Abstract

This paper evaluated the bibliometric study by Li et al. (Int Urol Nephrol, 2024) on machine learning in renal medicine. Although the study claims to summarize the forefront trends and hotspots in this field, several key issues require further clarification to effectively guide future research. Firstly, while the authors used the "*" wildcard to broaden the search scope, they screened articles only by document type and language, without specific filtering based on titles, abstracts, or full texts. This approach may have led to the inclusion of irrelevant studies, potentially compromising analytical accuracy. Secondly, the authors conducted the search using the Topic (TS) field, which may include articles not closely related to the intended topic. We recommend using Title (TI), Abstract (AB), and Author Keywords (AK) as filtering criteria in future studies to improve search precision. Finally, in the keyword co-occurrence analysis, the authors did not merge synonyms, leading to distortions in keyword frequency rankings; for example, "machine learning" and "machine learning (ML)" were treated as separate terms. We believe that synonym merging would enhance the accuracy of keyword analysis. Overall, the search strategy by Li et al. demonstrates issues such as imprecise scope and lack of synonym integration. To ensure the comprehensiveness and accuracy of future research, we suggest refining the search strategy, employing precise screening steps, and integrating synonyms to improve the quality of bibliometric studies.