发明名称 AI for relating herbal ingredients to illnesses classified in traditional chinese medicine/TCM using probabilities and a relevance index
摘要 Described herein are systems and methods for identifying herbal ingredients effective in treating illnesses in Traditional Chinese Medicine (TCM) using an artificial neural network.
申请公布号 US9275327(B2) 申请公布日期 2016.03.01
申请号 US201113807576 申请日期 2011.07.28
申请人 HERBMINERS INFORMATICS LIMITED 发明人 Lin Wilfred Wan Kei;Wong Jackei Ho Kei;Wong Allan Kang Ying
分类号 G06F17/00;G06N3/02;G06F19/00;G06F19/18 主分类号 G06F17/00
代理机构 Saliwanchik, lloyd & Eisenschenk 代理人 Saliwanchik, lloyd & Eisenschenk
主权项 1. A system for identifying at least one herbal ingredient effective in treating an illness classified under TCM (Traditional Chinese Medicine) comprising: one or more computer readable storage media; program instructions for a relevance index program module stored on the one or more computer readable storage media that, when executed by a processor, direct the processor to: construct TCM illness entity artificial neural networks from a plurality of TCM illness entities, wherein each TCM illness entity artificial neural network identifies a single TCM illness entity, wherein each TCM illness entity artificial neural network computes a relevance index indicating a probability of intersection between the single TCM illness entity and a selected one or more TCM herbal ingredient entities;construct TCM herbal ingredient entity artificial neural networks from a plurality of TCM herbal ingredient entities, wherein each TCM herbal ingredient entity artificial neural network identifies a single TCM herbal ingredient entity, wherein each TCM herbal ingredient entity artificial neural network computes a relevance index indicating a probability of intersection between the single TCM herbal ingredient entity and a selected one or more TCM illness entities;train the TCM illness entity artificial neural networks and the TCM herbal ingredient entity artificial neural networks from a first training dataset, comprising the plurality of TCM illness entities and one or more illness-associated-symptoms for each TCM illness entity of the plurality of TCM illness entities, and a second training dataset, comprising the plurality of TCM herbal ingredient entities and one or more ingredient-effective-symptoms for each TCM herbal ingredient entity of the plurality of TCM herbal ingredient entities;in response to receiving an indication of a referential entity selected from the plurality of TCM illness entities: computing relevance indexes for the referential entity using the TCM herbal ingredient entity artificial neural networks; andidentifying at least one effective TCM herbal ingredient entity from the relevance indexes.
地址 Hong Kong CN