≪脾胃論≫에 사용된 처방의 본초 네트워크 분석 (Network Analysis of the Herbs in Prescriptions from “Biwiron (脾胃論)”)

서지정보

· 발행기관 : 대한한방내과학회
· 수록지 정보 : 대한한방내과학회지 / 45권 / 6호 / 1195 ~ 1206페이지
· 저자명 : 최새롬, 신재욱, 박상민, 백지훈, 이민주, 최지연, 권태경, 이은지, 변정원

초록

Objective: This study aimed to analyze the network of herbs in “Biwiron (脾胃論)”, a text focused on digestive disorders, to identify the core herbs and the significant combinations that contribute to its prescriptions.
Methods: The prescriptions from “Biwiron (脾胃論)” were organized into a dataset listing each herb per prescription. A network model was created with herbs as nodes and their co-occurrences in prescriptions as edges. The analysis included frequency counts, centrality metrics (degree, closeness, betweenness, and eigenvector centrality), and community detection using the Girvan-Newman algorithm.
Results: Among the 59 herbs, Gancao (甘草) appeared most frequently. Chenpi (陳皮), Shenqu (神麴), and Baijiangcan (白殭蠶) scored highest in degree, closeness, and betweenness centrality, respectively, while Chenpi (陳皮), Shengma (升麻), and Huangqi (黃耆) topped eigenvector centrality. Based on the Leiden algorithm, the herbs analyzed were networked into five communities, which frequently appear together in various prescriptions and share similar efficacies and indications.
Conclusion: This study systematically uncovered the roles of core herbs and combinations in “Biwiron (脾胃論)”, highlighting the theoretical basis for understanding herbal interactions. Future studies could expand this analysis to broader datasets and clinical validation.

영어초록

Objective: This study aimed to analyze the network of herbs in “Biwiron (脾胃論)”, a text focused on digestive disorders, to identify the core herbs and the significant combinations that contribute to its prescriptions.
Methods: The prescriptions from “Biwiron (脾胃論)” were organized into a dataset listing each herb per prescription. A network model was created with herbs as nodes and their co-occurrences in prescriptions as edges. The analysis included frequency counts, centrality metrics (degree, closeness, betweenness, and eigenvector centrality), and community detection using the Girvan-Newman algorithm.
Results: Among the 59 herbs, Gancao (甘草) appeared most frequently. Chenpi (陳皮), Shenqu (神麴), and Baijiangcan (白殭蠶) scored highest in degree, closeness, and betweenness centrality, respectively, while Chenpi (陳皮), Shengma (升麻), and Huangqi (黃耆) topped eigenvector centrality. Based on the Leiden algorithm, the herbs analyzed were networked into five communities, which frequently appear together in various prescriptions and share similar efficacies and indications.
Conclusion: This study systematically uncovered the roles of core herbs and combinations in “Biwiron (脾胃論)”, highlighting the theoretical basis for understanding herbal interactions. Future studies could expand this analysis to broader datasets and clinical validation.

참고자료

· 없음