A Structured Coding Model And Fast Mapping Method For Chinese Clinical Terminology

With more and more heterogeneous medical information systems implemented in hospitals by different vendors, a mass quantity of name-and-coding-heterogeneous clinical terminologies have become the bottleneck of clinical information systems’ integration, data exchange and information sharing. So far, corresponding Chinese clinical terminology standards and systems are still in a primary stage and the efficient mapping tools are lacking correspondingly. And due to language barriers, the well developed international clinical terminology systems and mapping tools can’t support the specific tasks of Chinese clinical terminology retrieval and mapping.In this thesis, in order to solve the problem of mapping massive, diverse and heterogeneous Chinese clinical terminologies, we designed a Chinese clinical terminology coding model based on international mainstreaming medical terminology standard coding systems. Combined with Chinese word segmentation and text relevancy algorithms, we developed a mapping method for Chinese clinical terminology based on this structured coding model to map the terms accurately and rapidly. We also developed a tool for the proposed method, to provide an efficient and convenient way to reduce manual workload and shorten mapping time of Chinese clinical terminology mapping.We used the terms of drugs, medical imaging examinations and laboratory test items, collected from the set of National Health Industry Standards and clinical information systems used in two hospitals, to evaluate the proposed coding model and mapping method. The results show that the proposed coding model is stable, broadly applicable and scalable, which can significantly improve the accuracy of term mapping. The proposed method can quickly map heterogeneous terms in specific categories with high accuracy.