Latent semantic diagnosis in traditional chinese medicine

Traditional Chinese Medicine (TCM) is the main route of disease control for ancient Chinese. Through thousands of years' development and inheriting, TCM is the most influential traditional medical system which lasts the longest time and used by the largest population. However, there are still much space for data driven TCM information process to take advantage of for real medical application. In this paper, we propose a statistical diagnosis approach to find out the pathogenesises based on the latent semantic analysis of symptoms and the corresponding herbs. We assume that the latent pathogenesis is the inherent connection between symptoms and herbs within a medical case. We therefore develop a novel multi-content model based on LDA. Then three prescription recommendation algorithms are proposed focusing on permanent cure, symptom alleviation and both. We used the proposed model to analyze two TCM domains amenorrhea and lung cancer. Experiment results illustrate that the pathogenesises found by our model correspond well with the theory of TCM and the proposed model provides a theoretical data-driven way to establish diagnosis standards. And the prescription recommendation algorithms help doctor make treatment more accurately, which can lead the development of diagnosis of TCM.