Digitising Ancient Wisdom: AI Maps Traditional Chinese Medicine for Eye Health
Source PublicationInternational Journal of Ophthalmology
Primary AuthorsSchool of Chinese Medicine, Hunan University of Chinese Medicine, Changsha 410208, Hunan Province, China, Xiao, Wang et al.
Researchers have successfully bridged ancient medical practices and modern data science to tackle diabetic retinopathy (DR), a severe eye condition caused by diabetes. The team constructed a comprehensive 'knowledge graph'—a structured digital web that maps relationships between data points—specifically for Traditional Chinese Medicine (TCM) approaches to this disease.
To build this system, the study utilised high-quality literature and electronic medical records as raw materials. Using a specialised AI model to recognise medical terms, and assisted by human experts to verify connections, they organised the data into a searchable network. This network, stored in a database called Neo4j, comprises 822 specific nodes (concepts) and over 1,300 relationships.
The result is more than just an archive; it enables digital 'dialectical reasoning', mimicking the logic of TCM practitioners. When tested, an intelligent question-answering tool built upon this graph achieved a remarkable 95% precision and recall. This semi-automatic scheme not only preserves medical heritage but also provides a high-tech tool for managing patient health.