With the improvement of China's technological innovation capability, the domestic information industry has been developed vigorously, forming a unique value chain and supply chain in the market. Domestic operating systems, databases and security products are important technical products to promote the development of informatization and the process of informatization in various industries, and play very important roles in the current process of informatization management. In order to sort out the current development status of the domestic software industry, this article overviews the operating systems, databases, and security products in domestic software industries, and analyses the challenges and future development of domestic information industry.

Based on the concept of clinical decision-making support system, this paper reviews the development of clinical decision-making support system and its application and research progress in TCM field. This paper summarizes the connotation of TCM clinical decision-making support system, focuses on introducing its construction methods and research application, and introduces the TCM clinical decision-making support system which uses AI technology such as TCM natural language processing to conduct data-mining of TCM diagnosis and treatment and generate the TCM disease knowledge graph after modeling. This paper introduces the research progress and application of TCM clinical decision making support system in detail in order to provide new ideas for the research and application in the same field.

Objective By building a high-compatibility data lake based on the big data platform, to achieve the purpose of reducing data storage costs, reducing comprehension deviations on data and improving data quality. Methods By analyzing the actual operation and data application needs of the hospital, and based on the technical capability of the data lake base, the medical big data platform was improved and a full-volume medical data into the lake system was established to realize the storage, aggregation, resource utilization and capitalization of massive data. Results The construction and application of the medical data lake can help data governance, realize efficient and rapid access and processing of data, and provide data basic support capabilities for medical operations, scientific research, medical quality control, clinical services, decision-making management, etc. Conclusion According to the actual situation of hospital data and application, a convenient and efficient data governance approach is provided to improve the hospital's medical ability and enrich the service capabilities.

Objective To explore the method of high-precision information extraction and structured processing of electronic medical record by adopting natural language processing, structured algorithm and knowledge graph. Methods By constructing named entity recognition model, relation recognition model and synonym recognition model, the in-sentence information was extracted from medical record. A medical record spanning tree algorithm was proposed, which can effectively analyze the hierarchical structure of large paragraph medical record text. At the same time, knowledge graph technology was used to store the construction models of information extraction and hierarchical analysis to realize high-precision extraction of medical record text information. Results A set of high-precision medical  record extraction methods integrated with deep learning algorithm and structural analysis algorithm were formed, in which the accuracy of entity recognition model was 95.74%, and the accuracy of relationship recognition model was 89.20%. Finally, a structured medical record with clear hierarchical structure was generated, which can accurately locate and extract information. Conclusion The high-precision extraction method of medical record information explored in this paper integrates deep learning algorithm with structured analysis algorithm, takes into account the information extraction in sentences of medical record text and the analysis of medical record structure level, which can realize the automatic extraction, precise positioning and efficient management of medical record data, lay the data foundation for clinical medical research and also provide methodological reference for the mining of medical record text data of other diseases.

Objective To construct a knowledge graph of evidence-based guidelines of Chinese medicine based on PICO model, and to explore the use of this knowledge graph as a knowledge base to build an intelligent Q&A system to enhance clinical decision support. Methods The concept layer and data layer of the knowledge graph was designed by integrating top-down and bottom-up approaches. Taking polycystic ovary syndrome as an example, the “Wenxinyiyan” large model tool was used to extract knowledge, and the extraction results were stored in the Neo4j graph database, which realized the visual display of the knowledge graph. At the same time, clinical expert questionnaire survey and framework structure design of intelligent Q&A system were carried out. Results The knowledge graph of evidence based TCM guidelines containing 13 categories, 248 nodes, 12 relationship types, 337 relationships and 10 data attributes was constructed, and its visualization function and application examples have been demonstrated. In the part of the intelligent Q&A system, the guidelines have been divided into 12 major categories based on their clinical question ranges, and 24 kinds of question templates have been preset, which has laid the foundation for the subsequent question pre processing and provided a strong support for the system implementation. Conclusion This study ensures the scientific  and accurate clinical assistant decision-making system by using the evidence-based guidelines as the data source, and also provides new ideas and methods for intelligent information service in the field of traditional Chinese medicine.

Objective To strengthen the foundation of hospital information security, enhance the ability of self-control, carry out the IT-application innovation transformation of hospital information system and realize full scenario application. Methods According to the relevant principles of compliance, mainstream, feasibility and high performance, the technical route and products of IT-application innovation such as server, storage, terminal, operating system and database were selected. Based on the API conversion technology of Wine application container and the differences between the original database and the domestic database, the domestic operating system database were adapted to the hospital application software. Results After some adaptation and transformation were made to the application software under Windows and Oracle Environment, 120 in-use application software and domestic operating system's adaptation verification and application have been completed, and 89 in-use application software and domestic database adaptation verification and application have been completed within 18 months and 6 months respectively, greatly reducing the cost of application software transformation and the difficulty of database migration. Conclusion Under the premise of ensuring user experience, the appropriate IT-application innovation ecology can realize the full scenario application of the hospital full-stack localization platform with the fastest speed and the lowest cost.

Objective Based on the situation of the Class-III independent deep learning SaMD (Software as a Medical Device) and related enterprises in AI medical devices, to explore characteristics and intrinsic reasons of the two and to provide reference for industrial development. Methods Based on the public information obtained from the website of the State Drug Administration and the website of the National Medical Products Administration (NMPA) and the National Enterprise Credit Information Publicity System, products and enterprises information on Class-III deep learning SaMD was collected, and descriptive statistical analysis was conducted. Results Since 2020, 53 Class III deep learning SaMD have passed registration and approval in China, and the number of reviews has increased year by year, mainly in the lung and heart specialties. These products belong to 26 enterprises respectively, and the enterprises show a certain centralized trend in terms of establishment time and geographical distribution. Conclusion The sustainable development of AI medical devices needs to be guided by clinical applications continuously, strengthen supporting policies and put emphasis on the innovative role of technology and talents.

By analyzing the related concepts of Internet medical service at home and abroad, this paper systematically expounds the connotation, correlation and differences of related concepts such as telemedicine, Internet hospital, Internet diagnosis and treatment, tracks the origin and development, and summarizes the management enlightenment of developing high-quality Internet medical services in China′s public hospitals. From the perspective of bibliometrics, this paper analyzes 9 concepts in 4 categories, including generality, service model, service platform and service mode. Afterwards, the evolution characteristics are analyzed through literature research methods, so as to put forward suggestions for the high-quality development of Internet medical service in China. With the help of rapid development of information technology, Internet medical service has developed rapidly in China and has entered a stage of leapfrog development. China should further vigorously promote the development of public Internet hospitals and the deep integration of online and offline services, and transform from product-oriented to service-oriented to achieve high quality development of Internet medical service.

The purpose of this paper is to explore the application and construction of the knowledge base of ancient Chinese medicine books in clinical assistant decision-making system. Ancient books of Chinese medicine contain rich clinical experience and theoretical knowledge of Chinese medicine, however, due to the large number of ancient books, scattered distribution and uneven preservation status, it is difficult for doctors to fully utilize them in clinical practice. Therefore, integrating the knowledge base of ancient TCM books with clinical assistant decision-making support system has gradually become one of the important trends to improve the level of TCM clinical diagnosis and treatment. In this study, a total of 278 ancient books of Lingnan Chinese medicine were collected, covering 488 diseases, 1,198 signs and syndromes, and 3,907 symptoms. The ancient books of Chinese medicine were systematically organized and classified to construct a structured and standardized library of ancient books of Chinese medicine. At the same time, with the introduction of the modern clinical medicine related knowledge, a comprehensive clinical knowledge base of Chinese medicine was formed. In this study, a set of rule-based inference engine and machine learning algorithm were designed to retrieve the corresponding diagnosis and treatment suggestions from the knowledge base of Chinese medicine books based on patients’ symptoms, constitution and other information. Through practical application and testing, it is verified that the clinical assistant decision-making system based on the knowledge base of ancient Chinese medicine books can improve the effectiveness of TCM diagnosis and treatment, and the construction of the system not only helps to pass on and carry forward the knowledge of ancient Chinese medicine books, but also provides a strong support for modern clinical practice of Chinese medicine.

Objective Based on the TCM diagnosis and treatment concept of "state-targeted syndrome differentiation", this study aims to utilize AI technology to provide intelligent support for TCM assistant diagnosis and treatment of metabolic diseases. Methods By using natural language processing technology, information extraction and entity identification were carried out from medical literature and expert experience, and a terminology knowledge base containing disease states, therapeutic targets, and related syndrome differentiation rules was constructed. On this basis, the graph database and deep learning algorithm were utilized to construct the knowledge graph of the state-target syndrome differentiation, so as to express the complex relationship between different disease states and therapeutic targets. Results Through model training and knowledge reasoning and based on the specific symptoms and constitution of the cases, the corresponding TCM treatment plan was recommended, which improved the accuracy and efficiency of the assistant diagnosis and treatment of metabolic diseases. Conclusion This study provides new methods and ideas for TCM syndrome differentiation and treatment, continuously iterates and optimizes the algorithm, expands clinical applications, promotes the development and improvement of TCM assistant diagnosis and treatment, and lays a good technical foundation for TCM inheritance and innovation.

With the adjustment of the pneumonia prevention policy, the amount of chest CT detection has increased rapidly, and head hospitals and primary medical institutions are facing problems such as insufficient medical resources and insufficient diagnosis and treatment services. In order to carry out medical information research such as assisted screening or diagnosis in the field of medical imaging, many medical IT application staff have introduced computer vision and other related technologies to create a CT-assisted chest screening system for pneumonia. Taking the application of Artificial Intelligence (AI) chest CT intelligent-assisted screening application as an example, the application pattern and effect of AI-assisted diagnosis during the pneumonia pandemic, including COVID-19. It highlights the benefits of the system to improve the efficiency of clinical and radiological staff in image reading. Based on the accuracy and application efficiency of the AI-assisted diagnosis model, the efficacy of the AI-assisted diagnosis model in qualitatively judging the type of pneumonia and quantitatively estimating the area of the lesion was illustrated. Finally, based on the practical work of clinical research, the limitations and future prospects of the application are discussed.

Objective To reduce a series of problems such as clinical nurses' untimely measurement of patients' physical signs and inaccurate contents of nursing documents, and to facilitate clinical nurses in collecting the daily vital signs data such as body temperature and pulse of patients. Methods A set of vital signs assistant acquisition system integrated with hospital nursing information system was developed to automatically obtain personal information such as patient's medical orders, conditions and surgical records, which were compatible with nursing norms. The type, time and frequency of daily signs measurement were obtained intelligently by algorithm and the operator was prompted in time. Results After the applicaiton of this system, the problems of irregular nursing operation and untimely filling of nursing documents by clinical nurses were significantly reduced, the length of time for nurses to complete the nursing document records per day was significantly shortened, and the completeness rate of nursing documents in the department was significantly improved, with statistically significant differences (P＜0.05), which greatly improved the efficiency and the quality of nursing work. Conclusion The intelligent vital signs assistant acquisition system can improve the integrity and timeliness of nursing writing, with good user experience.

Objective To analyze the development status and characteristics of multi-campus hospitals, establish a multi-campus homogenized management information platform by means of IT, and achieve continuous improvement of medical service quality. Methods Through literature research, the concept and current research status of multi-campus hospitals were sorted out. Based on the integrated information platform and with the help of system and network transformation, the "two hospitals and three campuses" were interconnected, so as to realize the homogenization and intelligence of medical treatment and management. Results The construction of an information platform for homogenized management in multi-campus hospital was built, and the network architecture of multi-campus hospital was reconstructed, and the information sharing in multi-campus hospital with medical quality as the core. Conclusion The establishment of the information platform for homogenized management in multi-campus hospital is conducive to improving medical efficiency, providing patients with better medical services, promoting the construction of smart hospitals, and facilitating the high-quality development of hospitals.

Objective To design an online and offline integrated intelligent management platform for children's growth and development in accordance with the construction standards of smart hospitals and taking into account the characteristics of children's growth and development specialty. Methods The innovative application of Internet, AI and big data were used to realize the "Internet +" healthcare service of children growth and development, AI bone age assessment, and the big data intelligent assessment of children's growth and development. Results The platform provided dynamic monitoring, accurate identification and personalized guidance, and realized early detection, early intervention, early diagnosis and early treatment of children's growth and development deviations. Conclusion The establishment of the platform can improve the efficiency and quality of diagnosis and treatment, enhance patients'medical experience, and provide reference and inspiration for promoting the healthy development of children and improving the level of children's health management.

Objective To construct a knowledge graph based on the characteristics of Traditional Chinese Medicine (TCM) Daoyin (Chinese traditional body building exercise), so as to facilitate users to inquire, learn, spread traditional culture of TCM health care, and improve health quality. Methods The BERT-CRF model and manual proofreading were used to identify the named entities in literatures related to Daoyin practices, and the relationships between entities were established based on the disciplinary characteristics, and a knowledge graph of TCM Daoyin was constructed by using Neo4j. Results A total of 3,152 literatures were included. The constructed knowledge graph contains 2,262 entity nodes, 5,108 relational data, 7 entity category attributes, and 7 types of relationships. The knowledge graph can be retrieved by using the Cypher query language of Neo4j. Conclusion This knowledge graph can visually display the knowledge of TCM Daoyin, which can be used to build application programs and network platforms such as Daoyin knowledge retrieval, intelligent answering Q&A, and exercise recommendations. It can be used for developing relevant AI application and devices and for research and innovation in computer science and automation fields, and can also be used for in-depth data mining to discover new research directions.

Objective To understand the basic situation of personnel for IT-application in Hubei Traditional Chinese Medicine Hospital, and to propose countermeasures and suggestions for the construction of talents for IT-application in Chinese medicine hospitals. Methods Based on the survey of the status quo of IT-application in Hubei Provincial Traditional Chinese Medicine Hospital, SPSS 26.0, Excel and other software were used for statistical analysis by Chi-square test and Fisher exact probability method. Results The organization and position setting of the IT department in Hubei Provincial TCM hospital were basically complete, and the educational level and title distribution of IT-application talents' were gradually improving, indicating that the quality of IT-application staff was steadily improving, but there still existed a series of problems such as insufficient quantity, uneven distribution, unscientific ratio of academic qualifications and specialties, as well as lack of scientific research ability due to being occupied in daily maintenance. Conclusion It is suggested to strengthen the cultivation of IT-application talents in TCM hospitals, optimize the professional structure and talent ratio, improve the quality of talents, strengthen the cultivation of network security awareness, highlight the advantages and the characteristics of Chinese medicine, and strengthen the support of information technology to the Chinese medicine medical service system.

This paper introduces the construction of an ancient and modern literature retrieval system of Traditional Chinese Medicine (TCM) for vertigo disease from the aspects of data sources, construction methods, technical architecture, and functional modules. This system comprehensively collects and sorts out ancient books and some modern periodical literatures related to the etiology and pathogenesis of vertigo, syndrome classification, and prescription treatment. It can be used for rapid data mining, analysis, and visual display of vertigo literature data, and can realize the intercommunication and retrieval of vertigo-related ancient and modern literature, which can provide rich original materials for clinicians and researchers to study, and provide reliable data and information support for further analysis and research on the characteristics of vertigo literature.

Objective In order to consolidate the achievements of "Internet+ Healthcare" demonstration area construction in Ningxia Hui Autonomous Region (hereinafter referred to as Ningxia), and further improve the ability of integrating medical service and elderly care. Methods Based on the current situation of local medical and nursing services in Ningxia, big data and Internet technology were applied to construct an information system of comprehensive medical service and elderly care based on Internet healthcare, and an information platform of integrated medical service and elderly care was built for institutions at different levels. Results A multi-directional collaborative mechanism was provided for medical service and elderly care institutions, social elderly care institutions and home-based elderly population, forming a whole-process medical service and elderly care management model covering online and offline, which could provide all-round, multi-level and continuous medical service and nursing for the elderly. Conclusion The Internet+ medical service and elderly care integrated management system can realize the whole-process management of medical service and elderly care integration and interconnection, which is an active exploration of the "Internet +medical service and elderly care" integrated service model.

With the wide application of emerging technologies in medical and healthcare industry, the value of healthcare data is highlighted, but security challenges also rise. How to realize secure, autonomous and controllable cryptographic applications in medical and healthcare industry and promote the establishment of a unified standard for commercial cryptographic security applications have become an important issue in the field of medical and healthcare security. By taking Nanjing EMR Document Management System as the practice carrier, Nanjing Municipal Health Information Center made use of domestic commercial cryptography technology to modify the system from the perspectives of hardware, network, equipment, application data, key management and deployment, etc, established the security support system for commercial cryptographic application and passed the level-three security assessment, which can provide reference for the data security of health institutions and the transformation of domestic commercial cryptographic application.

As an important part of the operation and management of public hospitals, the fine management of human resources can not only improve the comprehensive quality and competitiveness of employees, but also improve the overall operational efficiency and service quality of hospitals. Starting from the construction of medical resource center focusing on the core business of medical staff business, position staffing, department structure optimization and professional structure optimization, this paper introduces the overall positioning and index analysis design of the medical human resources platform, and elaborates on the core functions of the system, such as model management, assessment plan management, and evaluation result management. Finally, the application practice is presented from the system level. Supported by data, the overall project can promote sustainable development of disciplines, rational utilization of personnel, efficient operation of business, high-quality development and digital transformation of hospitals.

To solve the problem that the general Tibetan pre-training language model cannot adapt to the processing of Tibetan medicine data and there is word information loss, this study proposes a method that integrates the lexical features of Tibetan medicine with the character features of the general Tibetan pre-training model, so as to improve the recognition of proper terms of Tibetan medicine and enhance the model's understanding of Tibetan medicine. This method constructs a feature dictionary in the field of Tibetan medicine, uses the dictionary to obtain the potential matching word set of each word in the training data, and then embeds word set features into the character representation to enhance the representation of Tibetan medicine characters. Experiments show that only using the general Tibetan pre trained model will reduce the performance of Tibetan medicine entity recognition. After integrated with the features of the word set, the F1 value is significantly increased by 17.19%. In addition, it is verified that this method cannot only supplement the lack of vocabulary information in the pre-trained model, but also effectively alleviate the problem of mismatch between the model and Tibetan medicine data.

Large language modeling (LLM) has powerful learning and reasoning ability, but due to the lack of internal knowledge, the model faces hallucination challenges. To address the need for high-quality professional knowledge in medical Q&A, this study proposes a medical Q&A method by integrating LLM with knowledge graph, and applies it to Q&A system construction of Medical Science and Technology Information Knowledge Service Platform (MedKaaS). The study shows that compared with the traditional knowledge base Q&A (KBQA) and the emerging Q&A based on generative LLM, the method of integrating LLM with knowledge graph proposed in this paper can fully synthesize the advantages of both, provide evidence traceability support for medical Q&A, and effectively improve the quality of medical intelligent Q&A and the effect of medical knowledge service.

Objective To address the low efficiency of manual screw layout planning during robot-assisted orthopedic surgery, and to achieve automatic, efficient, and high-quality layout planning of vertebral pedicle screws based on CT images. Methods A supervised machine learning method was used with a deep learning neural network to segment and layout screws for vertebral pedicles using CT scans. 440 screws from 44 cases of lumbar spine CT scans were used as training sets, and 110 screws generated from 11 cases were used as testing sets. The results of manual layout planning were control group, and blinding was used in the experts' evaluation for the layout planning, while the layout duration was recorded for the planning efficiency evaluation. Results The clinical availability rate of the automatic layout planning method was 95.4%, and the average automatic layout planning duration and manual planning duration were 68.8 seconds and 177.6 seconds, respectively. Conclusion This method can initially achieve efficient and high-quality automatic layout planning of vertebral pedicle screws. However, clinician supervision and review are still necessary.

Objective Due to the privacy of children with autism, their datasets are often difficult to obtain, which creates certain difficulties for related research. For this reason, this paper constructs a database of autistic children's characteristics. Methods Firstly, an interaction scene was designed and a facial dataset of children with autism was collected, and then facial visual features of children with autism were extracted using techniques of facial key point detection, line-of-sight tracking, head posture and facial action units in computer vision, and then all the feature data were encoded on the basis of the extracted facial features, so as to construct a database of visual features of children with autism. Results This visual feature database can be used by researchers in this field. In addition, based on this visual feature database, this paper analyzed and predicted children with autism using facial action units as an example, and the experimental results showed that the classification accuracy was up to 93%. Conclusion This database can be used in the research related to autistic children, and has a greater application value.