Workshop on Advances and Issues in Traditional Chinese Medicine Clinical Data Mining (AI-TCM)

In conjunction with the 15th Pacific-Asia Conference on Knowledge Discovery and Data Mining

24th May 2011, Shenzhen, China

Important Dates

  • Submission deadline 23:59, 14th January 2010 PST
  • Author notification 4 February 2011
  • Camera-ready due 18 February 2011

Motivation

Traditional Chinese Medicine, also known as TCM, has a long history of being considered as one of the major medical approaches in China. Different from its western counterpart, TCM embraces a holistic rather than a reductionist approach. While medications in western medicine often strive to avoid drug interaction, TCM works oppositely by exploiting the useful interaction among multiple herbs that are used in a formula (prescription). The efficacy of a TCM medication derives from the interaction of these herbs. Furthermore, while western medicine has only recently begun to popularize the prescription of personalized medicine according to a patient's genome, Chinese medical doctors have long been practicing this approach. In TCM practice, every prescription is a clinical trial from the viewpoint of western medical practice. In terms of clinical study, TCM clinical practice is a kind of real-world clinical trial, in which all the practical treatments aim to maximize the effectiveness. Therefore, the clinical data generated in the clinical operations are important to hypothesis generation and clinical guideline development.

Historically in TCM, all these knowledge were accumulated and formulated through experience rather than the support of a proper scientific research. But we have witnessed a significant change in recent years. There is currently a strong push in the globalization of traditional Chinese medicine and practice. This movement encourages the use of different techniques, such as statistical and data mining methods, to find the underlying biomedical principles and theories, and to unlock the secrets of this "ancient" conservative domain of TCM.

Massive amounts of TCM clinical data are now being collected at an increasing pace. Data mining can help unveil the hidden precious knowledge, such that it becomes a complementary tool to assist TCM clinical research. There are many aspects to TCM, e.g. theory discovery, therapy methodology analysis, symptom detection, prescription analysis, herbal drug design, and intelligent diagnosis. It is anticipated that the sharing of experiences will provide new ideas in research directions and provide the stimulus for creative breakthroughs. Furthermore, in recent years, many researchers in computer science and statistics have joined in the TCM data mining field.

Topics

Original papers on TCM clinical data mining are called for. The topics include but are not limited to the following:

  • Domain-driven data mining foundations.
    • Effectiveness-driven knowledge discovery
    • Integrative clinical data mining (integrate different TCM data sources)
    • TCM clinical literature-based discovery
    • Data preprocessing and data quality control
    • Novel Methodology in Bioinformatics
    • Novel Methodology in Systems Biology
    • Novel Methodology in Pattern Recognition
  • Applications in TCM fields (algorithms and software tools).
    • Theory Discovery and hypothesis generation.
    • Clinical Prescription (e.g. herbal medicine and acupuncture prescription) Analysis and Mining
    • TCM Diagnosis (syndrome differentiation)
    • The relationships between TCM entities (e.g. symptoms, syndromes and herbs) and modern biomedical entities (e.g. diseases, laboratory test results and physical tests).
    • Mining from real-world TCM clinical data
    • Mining from TCM clinical trial data
    • Mining from TCM scientific basic databases (e.g. herb database, formula database and disease database)
    • Herbal Drug Design