A team around LIU Zhi (CAS Shenyang) and QI Liang (Qingdao University of Science and Technology) has developed a neural-network based system to simulating a doctor’s inquiry and his prescription that is composed by a series of herbs. It can automatically simulate some principles and learns the interaction between symptoms and herbs from clinical records of traditional herbal medicine. This model consists of two different attention mechanisms for distinguishing the main symptoms and pays attention to different symptoms. In 22.41% of all cases tested, several herbs in each predicted prescription overlapped with its label; and 10.1% of cases wesre completely different from the label. In summary, 67.49% of the predicted prescriptions are close to the labels, and 89.9% among them contain the same herbs with the labels.
CAS team develops AttentiveHerb, a neural-network-based algorithm for TCM prescriptions