Acupoints for Headache with Blood Stasis Syndrome: A Literature Study Based on Data Mining Technology [Letter]

Dear editor

We read with great interest the article by Shi et al¹ entitled “Acupoints for Headache with Blood Stasis Syndrome: a Literature Study Based on Data Mining Technology”. Through rigorous data mining methodologies, this study offers a comprehensive and systematic analysis of acupoint selection in acupuncture therapy for blood stasis headaches, contributing novel insights and valuable references to acupuncture. However, upon scrutiny and discussion with our peers, we identify several aspects of the writing that warrant further exploration and refinement to enhance the accuracy and impact of research in this domain.

Firstly, although the article mentions screening literature from multiple databases, detailed descriptions of specific inclusion criteria, exclusion criteria, and methods for assessing the quality of included studies (eg, JADAD score) are lacking.² This may cast doubt on the reliability and representativeness of the study’s findings among readers. For future reviews of this nature, we recommend explicitly outlining the screening criteria and processes, along with the tools or standards employed to evaluate the quality of the included literature, thereby improving research transparency.

Secondly, while emphasizing the application of data mining techniques in revealing acupoint selection patterns, the article provides only brief details on the specific algorithms used, parameter settings, and model validation procedures. To better convince readers and ensure the reliability of results, we suggest a detailed elaboration on the data mining techniques, encompassing the principles of the algorithms (such as association rule mining and cluster analysis), parameter tuning processes, and both internal and external validation results of the models. Ideally, these details could be included as supplementary materials for readers’ reference. This will facilitate peer understanding, replication of studies, and ultimately, advance academic progress.

Furthermore, while the article provides a detailed analysis of frequently used acupoints and their combinations, particularly emphasizing the core combinations of Fengchi (GB20) with Taiyang (EX-HN5) and Baihui (GV20), there is ample room for improvement in interpreting the significance and implications of these findings. Indeed, acupuncture textbooks consistently list GV20, GB20, and Touwei (ST8) as primary acupoints for treating internal headache disorders.³ Therefore, it would be prudent for the authors to further explore whether the observed high-frequency combinations exhibit unique therapeutic advantages over the conventional wisdom established in textbooks and clinical guidelines. A comparative analysis could shed light on whether these combinations exhibit specificity in addressing blood stasis headache that goes beyond the standard protocols, potentially uncovering novel insights into acupuncture therapy tailored specifically for this condition. Such a study would not only validate the effectiveness of the discovered combinations but also contribute to the advancement of acupuncture practice by identifying potential new therapeutic strategies.

It is acknowledged that the article acknowledges some limitations and suggests future research directions. However, these suggestions appear broad. To better guide subsequent research, we propose refining these directions. For example, in expanding data sources, specifying potential international databases or research networks for collaboration would be beneficial. In terms of objective efficacy assessment, exploring modern biomedical technologies (eg, functional magnetic resonance imaging, electroencephalography) that may aid in more precise evaluation of acupuncture outcomes could be discussed.⁴

In conclusion, the article holds significant academic value and practical implications within the realm of acupuncture. However, through enhancements in refining literature screening and data representation assessments, clarifying the specific application and validation of data mining techniques, delving deeper into result interpretation, refining future research directions, and optimizing writing style and language expression, the quality and influence of the article could be further elevated. We eagerly anticipate the emergence of more high-quality research findings in this field.