The 19^th World Conference on Medical Informatics (MedInfo) hosted by the International Medical Informatics Association (IMIA) was held in Sydney, Australia from July 8 to 12, 2023. This prestigious international event brought together thousands of digital health leaders and practitioners at the forefront of medical health. DHCtech was invited to attend and give an oral academic report, which showed the huge achievements of DHCtech in promoting the development of biomedicine and health informatics. At the same time, four wall newspapers were presented, covering original technologies and research achievements such as research on privacy protection of Chinese medical data and research on standardization of Chinese medical data, which attracted the attention of participants.

【Academic Lecture】

In the plenary session and keynote speech on 10th, Dr. Gong Meng Chun, Senior Vice President and Chief Medical Information Officer of DHCtech, was invited to attend the meeting and give an oral academic report.

He introduced to international experts the data technology support and corresponding fruitful results provided by DHCtech in different fields such as kidney disease, malignant tumor, childhood immune disease, cardiovascular disease and rare diseases for clinical research, translational medicine research, real-world research and artificial intelligence, which showed the accumulated accumulation and rapid progress in the field of biomedical and health informatics (BMHI).

As a new interdisciplinary subject, BMHI is in full swing all over the world, which provides core intellectual support for the development of biomedical industry, the improvement of clinical diagnosis and treatment technology and the development of artificial intelligence research in life sciences.

【Poster Presentation】

At this conference, 4 posters of DHCtech covered original technologies and research results such as research on privacy protection of Chinese medical data and research on standardization of Chinese medical data.

Poster Session1

Improvement and thinking of manual annotation in medical natural language processing: A case study 

Research content: This paper studied the influence of manual annotation method of single-center entity on the semantic extraction results of Chinese electronic medical record text, classifies and splits different medical texts according to data sources, and compared the effect of manual annotation of single-center entity long text with multi-model based on big data and multi-center text data.

Research significance: NLP model trained by single-center entity can reduce the training time and response time of new requirements, improve the overall data quality of NLP model and reduce the occurrence of semantic ambiguity, so as to better handle and utilize a large number of text data.

Poster Session 2

Application of SNOMED CT in the standardization of Chinese disease names 

Research content: Based on the smallest semantic unit of SNOMED CT, Chinese disease diagnosis names were intelligently segmented by using natural language processing technology. Through the intelligent mapping of the training model corpus to the corresponding attribute value domain, based on the concept model, expression model and relationship model, the intelligent matching of SNOMED CT concepts was realized by using the complete entity comparison algorithm.

Research significance: SNOMED CT plays a vital role in the standardization of Chinese disease diagnosis names, which can obviously improve the accuracy and application scope and promote the sharing and utilization of medical information.

Poster Session 3

Innovatively apply concept model of SNOMED CT to DHC drug terminology

Research content: Drug terminology of DHCtech used the conceptual model of SNOMED CT to divide drug categories and Parent-Child hierarchy. By constructing a conceptual expression after the combination of attributes, a fully defined drug name was further generated.

Research significance: Drug terminology of DHCtech has played a potential role in improving the interoperability of various electronic drug systems, and can form data sharing and integrated analysis among different hospitals.

Poster Session 4

A de-novo Multiple Hierarchical Ontology System for Herbal Medications in Traditional Chinese Medicine

Research content: The hierarchical features of SNOMED CT were used to construct a Chinese herbal medicine coding system, which covered the important functions of guiding clinical practice and generating real evidence, and defines the main attributes and relationships. The information extraction process was established by using natural language processing method to verify the performance of the new system in the standardization of traditional Chinese medicine information.

Research significance: The new multi-level Chinese herbal medicine coding system provides a feasible method for extracting useful information from Chinese medical records and literacy, which is an important way for evidence generation and modernization of Chinese medicine.