Deep learning in traditional Chinese medicine

Area of Science:

• Integrative Medicine
• Computational Biology
• Pharmacology

Background:

• Traditional Chinese Medicine (TCM) relies on ancient knowledge, necessitating modern scientific validation for therapeutic translation.
• The increasing volume of data from TCM experiments, clinical practice, and literature presents opportunities for advanced analysis.
• Deep learning (DL) has emerged as a key technology for data mining in various scientific fields, including TCM.

Purpose of the Study:

• To review recent applications of deep learning (DL) in traditional Chinese medicine (TCM) research.
• To highlight the potential of DL in extracting meaningful information from complex TCM datasets.
• To identify challenges and future directions for DL in TCM studies.

Main Methods:

• Literature review of recent studies employing DL techniques in TCM.
• Categorization of DL applications across different TCM research areas: medical image processing, medicinal substance investigation, data fusion, and natural language processing.
• Analysis of the effectiveness and limitations of DL methods in handling TCM data.

Main Results:

• DL has been successfully applied in medical image analysis, drug discovery, data integration, and text mining within TCM.
• DL methods demonstrate significant potential for uncovering hidden patterns and extracting valuable information from TCM data.
• The efficacy of DL is contingent upon the availability and quality of TCM-specific datasets.

Conclusions:

• Deep learning presents a promising avenue for advancing traditional Chinese medicine research through sophisticated data analysis.
• High-quality, curated TCM data are crucial for the continued development and successful implementation of DL techniques.
• Future research should focus on generating robust datasets to fully leverage the capabilities of DL in understanding and applying TCM principles.