Zhongguo Zhong Yao Za Zhi. 2025 Oct;50(20):5882-5895. doi: 10.19540/j.cnki.cjcmm.20250708.501.
ABSTRACT
Based on the ancient books, data mining of ancient medical records of epilepsy was carried out. The law of syndrome differentiation and treatment of epilepsy in ancient medical records was summarized, and effective prescriptions and core formulas were excavated, so as to provide reference for clinical treatment of epilepsy. Apriori algorithm module in IBM SPSS Modeler 18.0 was used to analyze the association rules of syndrome elements in ancient medical records of epilepsy. Fruchterman-Reingold network layout algorithm in Gephi 0.10.1 software was utilized to analyze the complex network of syndrome elements in ancient medical records of epilepsy, and the Pearson correlation coefficient in the measuring interval of SPSS Statistics 29.0 was employed to perform hierarchical cluster analysis on syndrome elements in ancient medical records of epilepsy. The law of syndrome differentiation and treatment was summarized, and effective prescriptions were explored. Through the statistical analysis of 117 ancient medical records of epilepsy, 22 disease nature syndrome elements, 10 disease location syndrome elements, and 4 emotional syndrome elements were extracted. Seventy-seven ancient medical records of epilepsy during the seizure period were included, with 71 medical records containing disease nature, 49 medical records containing disease location, and 11 medical records containing emotional elements. Twenty-nine ancient medical records of epilepsy during the resting period were included, with 23 medical records containing disease nature, 17 medical records containing disease location, and 6 medical records containing emotional elements. Through the compatibility analysis of syndrome elements, 6 association rules of disease location-nature during the seizure period, 13 association rules of disease location-nature, and 3 association rules of disease location-nature during the resting period were obtained. The Fruchterman-Reingold network layout algorithm in Gephi 0.10.1 software was used to visually analyze the syndrome elements in 70 medical records, disease location and nature in 49 medical records during the seizure period, and disease location and nature in 15 medical records during the seizure period. A total of 239 traditional Chinese medicines(TCMs) were obtained. The Pearson correlation coefficient in the measuring interval of SPSS Statistics 29.0 was used for linkage cluster analysis of TCM administered during the seizure period(frequency≥10) and the resting period(frequency≥4) in ancient medical records of epilepsy. With a cluster spacing of 24, the seizure period was divided into 2 groups, and the resting period was divided into 2 groups. The law of syndrome differentiation and treatment of epilepsy was that the liver-gallbladder fire was excessive, and the phlegm-fire disturbed the nerves in the whole process of epilepsy. The treatment was mainly resolving phlegm and soothing the nerve. The seizure period was characterized by wind-phlegm disturbing the brain and blinding the brain. The treatment was mainly eliminating phlegm and clearing fire as well as dispelling wind and calming convulsion, and the core formula was a modified combination of Changpu Yujin Decoction combind Lingjiao Gouteng Decoction and Erchen Decoction. The resting period was characterized by Yin deficiency of the liver and kidney, spleen deficiency, and excessive phlegm. The treatment was mainly eliminating phlegm and clearing fire as well as replenishing deficiency and nourishing Yin, and the core formula was a modified combination of Congming Decoction combind Zhusha Anshen Pills.
PMID:41508219 | DOI:10.19540/j.cnki.cjcmm.20250708.501
