ABSTRACT
Objective:To explore the medication law and core Traditional Chinese Medicine (TCM) compounds in the treatment of blood stasis vascular dementia (VD) based on data mining.Methods:The literature about TCM treatment for blood stasis VD was retrieved from the databases of CNKI, Wanfang, VIP, and CBM from January 2000 to November 2021. Microsoft Office Excel 2019, SPSS Modeler 18.0, SPSS Statistics 25.0, R X64 4.1.2, and Origin 2021 were used to perform medication frequency analysis, frequency analysis of four properties and five tastes of TCM, association rules, clustering analysis, factor analysis and data visualization.Results:A total of 196 articles were included, with 196 TCM prescriptions, involving 200 kinds of Chinese materia medica. High-frequency drugs were for Acori Tatarinowii Rhizoma, Chuanxiong Rhizoma, Salviae Miltiorrhizae Radix et Rhizoma, Polygalae Radix, Carthami Flos. The medicinal properties were mainly warm, mild and cold, the tastes were mainly sweet, bitter and pungent, and the meridians were mainly liver meridian, spleen meridian and heart meridian. A total of 19 association rules were obtained from the analysis of association rules for 2 kinds of Chinese materia medica, and the rules of the representative were Acori Tatarinowii Rhizoma- Polygalae Radix, Chuanxiong Rhizoma- Carthami Flos, Acori Tatarinowii Rhizoma- Curcumae Radix. A total of 4 categories were extracted through clustering analysis. Factor analysis extracted a total of 8 common factors. Conclusion:The core pathogenesis of blood stasis VD is blood stasis blocking brain collaterals, and there were also pathological factors such as qi deficiency, yin deficiency, phlegm turbidity and so on. The basic treatment is promoting blood circulation and removing stasis, and different methods of promoting blood circulation and drugs are selected. The methods of strengthening spleen and reducing phlegm, nourishing yin and blood, inducing resuscitation, tonifying the kidney and spleen, regulating qi, promoting collaterals and so on can also be used based on syndromes and symptoms of the patients.
ABSTRACT
Advances in biological and medical technologies have been providing us explosive volumes of biological and physiological data, such as medical images, electroencephalography, genomic and protein sequences. Learning from these data facilitates the understanding of human health and disease. Developed from artificial neural networks, deep learning-based algorithms show great promise in extracting features and learning patterns from complex data. The aim of this paper is to provide an overview of deep learning techniques and some of the state-of-the-art applications in the biomedical field. We first introduce the development of artificial neural network and deep learning. We then describe two main components of deep learning, i.e., deep learning architectures and model optimization. Subsequently, some examples are demonstrated for deep learning applications, including medical image classification, genomic sequence analysis, as well as protein structure classification and prediction. Finally, we offer our perspectives for the future directions in the field of deep learning.