RÉSUMÉ
Purpose/Significance By integrating clinical and biological sample information,a big data research platform for biologi-cal sample information resources is built to provide one-stop data retrieval,integration and analysis services for researchers,and a data governance system is established,so as to improve the level of hospital clinical research infrastructure construction.Method/Process Common data model and data governance technology are adopted to integrate data sources from different vendors through extraction,trans-formation,loading and other steps to provide a unified data access portal.Result/Conclusion The big data research platform for biologi-cal sample information resources has the advantages of multi-dimensional data screening and rapid integrated analysis,which can pro-vide support for clinical research.
RÉSUMÉ
Electrocardiogram (ECG) is a low-cost, simple, fast, and non-invasive test. It can reflect the heart's electrical activity and provide valuable diagnostic clues about the health of the entire body. Therefore, ECG has been widely used in various biomedical applications such as arrhythmia detection, disease-specific detection, mortality prediction, and biometric recognition. In recent years, ECG-related studies have been carried out using a variety of publicly available datasets, with many differences in the datasets used, data preprocessing methods, targeted challenges, and modeling and analysis techniques. Here we systematically summarize and analyze the ECG-based automatic analysis methods and applications. Specifically, we first reviewed 22 commonly used ECG public datasets and provided an overview of data preprocessing processes. Then we described some of the most widely used applications of ECG signals and analyzed the advanced methods involved in these applications. Finally, we elucidated some of the challenges in ECG analysis and provided suggestions for further research.
Sujet(s)
Humains , Troubles du rythme cardiaque/diagnostic , Électrocardiographie/méthodes , AlgorithmesRÉSUMÉ
Objective It combines medical big data and machine learning techniques to explore clinical outcomes based clinical physician performance evaluation method.Methods The non-negative principal component analysis(NPCA)was used in cases.Based on the non-negative sparse principal component analysis(NSPCA),a comprehen-sive index fitting was performed on 11 clinical performance indicators of 170 clinicians treating cardiovascular diseases.At the same time,confidence intervals were constructed based on root cause assessment techniques to calculate the range of indicators for each clinician.Results The coincidence rate of outpatient discharge diagnosis,the rate of grade A healing of surgical incision,the proportion of surgical patients,the rate of 3-day diagnosis,the proportion of third-grade and fourth-grade surgery,the completion of surgery and the number of operations were significant in dis-tinguishing the work performance of clinicians.However,the average length of hospital stays before surgery,the rate of unplanned readmission within 30 days,the average length of hospital stays of discharged patients,the main diag-nosis and cure/improvement,and the number of patients admitted were not significant in distinguishing the clinical work performance of clinicians.The overall work performance of all clinicians can be ranked through comprehensive index fitting,and the further evaluation of high,middle and low performance of each specific index can reveal the potential reconstruction dimensions of each clinician.Conclusion It utilizes machine learning techniques to achieve a comprehensive evaluation of clinical performance,utilizing medical big data as the foundation.It holds the potential to provide important support for a more scientific and objective assessment of clinical performance.
RÉSUMÉ
Real World Data (RWD) has various types of data sources, but each source has a different format and terminology code, which makes analysis process cumbersome and repetitive. The OMOP Common Data Model (CDM) is an open standard for analysis of RWD on a global scale, and the OHDSI community is responsible for its maintenance and development. What sets the OMOP CDM apart from other data standards is the way in which it has created a structure for integrating and handling terminology globally, and the way in which analysis is conducted without exposing individual patient information outside. Such features facilitate international collaboration. The method of not releasing patient data outside is expected to be widely utilized in future because it is highly compatible with Japan's pseudonymously processed information (PPI) based on the personal information protection act, in which PPI data cannot be provided to any third party but the purpose of use can be easily changed. There are many advantages not only for international collaboration, but also for domestic collaboration or in-house use. Epidemiologists and data scientists will be able to handle data in the same model they are accustomed to both domestically and internationally. That will be of great benefit to students, personnel, and their organizations especially when they study abroad, return home, or transfer internationally. Globally, collaborators from more than 70 countries are working on this project. Data on more than 800 million people after eliminating estimated duplicates, or 10％ of the world's population, has been converted to the OMOP CDM. More than 250 related published articles have been registered with PubMed. On the other hand in Japan, there are many issues to be solved, such as support system and terminology mapping. To catch up with international levels, strong cooperation from a wide range of fields is needed.
RÉSUMÉ
Objective:To establish a IgA nephropathy (IgAN) standard dataset for the structured and standardization of IgAN clinical information, which will be beneficial to the integration and utilization of clinical information among different medical institutions. Therefore, the IgAN Expert Collaboration Group composed the "IgA Nephropathy Standard Dataset".Methods:Referring to the domestic information standards, guidelines, data standard and consensus of related fields, based on electronic medical history, the patient identification number was used as the primary key of the system to collect information. By standardizing each data element in the data set, the standardization of the management system in data and information exchange, data collaboration and sharing was ensured, and a quality control system was developed.Results:This standard dataset included 607 data elements and 8 business domains, which were patient information, medical history information, physical examination, laboratory examination, assistant examination, renal pathology, drug treatment, and follow-up, respectively. Each module was composed of module name, data element name, English name, definition, range, reference standard, etc. At the same time, a corresponding quality control system was formulated to evaluate data quality from multiple dimensions such as completeness, standardization, accuracy, timeliness, and security for ensuring the high quality and security of the data.Conclusion:The IgAN standard dataset is established, which will contribute to the structuration and standardization of clinical information of IgAN patients.
RÉSUMÉ
Objective:To analyze data safety problem raised from personal medical data sharing and privacy protection, provide suggestions for improving its application and development.Methods:The personal medical data sharing and privacy protection measures were discussed according to the study of related literatures, typical case analysis, analyzing the current situation and its development trend of legislation status of personal medical data sharing and privacy protection.Results:Medical data is one kind of personal data, but more sensitive than other personal data. The country should strengthen relevant legislation, more clearly define relevant concepts, establish the authority and scope of personal medical data processing, improve the effectiveness and operability of laws, maximize the public interest of data, and ensure the balance between the personal data processing and privacy security.Conclusions:The personal medical data sharing and privacy protection is a systematic project. The solution of the personal security risks is also based on a comprehensive safeguard system which including laws, regulations, management and technology.
RÉSUMÉ
Objective:To establish an evaluation index system for the refinement management of respiratory specialties in four dimensions: medical care, teaching, scientific research and personnel training, to develop a refinement management system based on medical big data, and evaluate the effectiveness of its application.Methods:Based on evaluation indexes and literature of several domestic respiratory specialty influence rankings, an index system covering four dimensions (primary indexes), 16 secondary indexes and 73 tertiary indexes covering medical treatment (30%), teaching (15%), scientific research (40%) and personnel training (15%) was devised using brainstorming, Delphi, and hierarchical entropy weighting. Data from 13 professional groups and 248 respiratory discipline members integrated by the system in 2018 and 2019 were statistically analyzed using the refined index system embedded in the system to understand the system's application after a year of use with comparing the four dimensions of the discipline and the changes among specialty groups before and after implementation.Results:In 2019, respiratory medicine′s comprehensive score grew 30% compared to 2018. The subscores also grew, with scientific research showing the largest growth rate of 136% and talent cultivation showing the second highest. In 2019, ten of the 13 professional groups grew by 2% to 135%.Conclusions:An improved management system for respiratory medicine based on big data technology has provided a comprehensive, intuitive and quantitative view of the discipline′s development in four dimensions and the differences among specialty groups. It will be an effective decision-making tool for ensuring high-quality and balanced discipline development.
RÉSUMÉ
With the rapid development of global information technology, the wide application of electronic medical record management system and the construction and operation of biological sample base platform, the research value of medical big data is increasingly valued.In recent years, as a bridge between basic medical research and clinical medical research, the progress of translational medicine has greatly shortened the space-time distance between basic and clinical medicine, and also opened up a new prospect for the application of medical big data.As a methodology of clinical medical research, epidemiology emphasizes the design, measurement and scientific evaluation methods of medical research, which plays an important role in every link of translational medical research and is an important part of translational medical research.It takes systematic epidemiology as the main research method, deeply excavates and makes full use of medical big data resources, promotes translational research and knowledge integration with the rapid development of big data, the clinical transformation of basic research results of ophthalmology has entered a new era, greatly promoting the cross integration of ophthalmology and related fields.
RÉSUMÉ
As a computer science that seeks to simulate the problem of human intelligence, artificial intelligence ( AI ) is developing rapidly in many fields. The application of AI in the field of ophthalmology is increasing. With the development of medical informatization and internet medical care, medical data and machine learning algorithms continue to accumulate, and AI systems are continuously optimized and upgraded during the development of technology and applications. This paper summarized the application status of AI in ophthalmology from the aspects of data demand,source and format,application and optimization innovation of related algorithms,demand and improvement of hardware computing force,and analyzed the development status,challenges and future directions. Although there are some problems to be solved in the current development and application of AI,it is believed that AI will play an important role in clinical medicine in the near future.
RÉSUMÉ
With the advent of a new era of digital medicine,the emergence of a series of digital intelligent diagnosis and treatment technology has played an important role in the development of diagnosis and treatment of biliary malignant tumors.In this paper,the application of digital intelligent technology in the diagnosis and treatment of biliary malignant tumors is elaborated based on the relevant literature at home and abroad as well as the author's practical experience in the past 15 years.This paper introduces the application of digital medical technology represented by three-dimensional visuali zation,virtual simulation surgery,three-dimension printing technology,virtual reality,abdominal surgery navigation,medical big data and artificial intelligence-radiomics in preoperative evaluation,operation planning,and real-time intraoperative guidance,and looks forward to the new direction of intelligent technology in the diagnosis and treatment of biliary malignant tumors,so as to promote its diagnosis and treatment mode to intelligent assisted diagnosis and treatment.
RÉSUMÉ
With the development of information technology and the arrival of the era of big data,our country has introduced a number of policies and regulations to guide the application and development of big data in many industries including health care.This article introduced the background and significance of the development of medical big data,reviewed the characteristics of foreign big data platforms,discussed the management and application of medical big data platform,and anticipated the future development of big data for gastrointestinal cancer and even the entire medical industry.
RÉSUMÉ
To introduce Medical Information Mart for Intensive Care (MIMIC) database and elaborate the approach of critically emergent research with big data based on the feature of MIMIC and updated studies both domestic and overseas, we put forward the feasibility and necessity of introducing medical big data to research in emergency. Then we discuss the role of MIMIC database in emergency clinical study, as well as the principles and key notes of experimental design and implementation under the medical big data circumstance. The implementation of MIMIC database in emergency medical research provides a brand new field for the early diagnosis, risk warning and prognosis of critical illness, however there are also limitations. To meet the era of big data, emergency medical database which is in accordance with our national condition is needed, which will provide new energy to the development of emergency medicine.
RÉSUMÉ
Medical practice generates and stores immense amounts of clinical process data, while integrating and utilization of these data requires interdisciplinary cooperation together with novel models and methods to further promote applications of medical big data and research of artificial intelligence. A "Datathon" model is a novel event of data analysis and is typically organized as intense, short-duration, competitions in which participants with various knowledge and skills cooperate to address clinical questions based on "real world" data. This article introduces the origin of Datathon, organization of the events and relevant practice. The Datathon approach provides innovative solutions to promote cross-disciplinary collaboration and new methods for conducting research of big data in healthcare. It also offers insight into teaming up multi-expertise experts to investigate relevant clinical questions and further accelerate the application of medical big data.
RÉSUMÉ
Depending on Shanghai medical big data center and taking the medical big data after quality control and before data utilization as research object,the paper establishes the data cleaning frame,gives the evaluation method for data availability,finds out the corresponding cleaning strategies according to the clustering analysis of data characteristics and repeatedly deduces the accuracy,reliability of the strategy,thus providing a strong support for the analysis and utilization of medical big data.
RÉSUMÉ
Objective To investigate the relationship between the trauma severity and the usage of antibacterial drugs and to provide reference for standard protocol of proper antibiotic use in wound care.Methods ICD-10 and AIS were used to set up the relationship and to analyze the use of antibiotics in patients with different trauma score.Results 25 035 trauma patients were enrolled in this study.Those patients were divided into five groups according to the AIS score with least severe as group 1 to most severe as group 5.The patient percentage in group 1 to 5 was 21.92％,67.73％,8.86％,0.97％ and 0.52％ respectively.The five most frequently used antibiotic classes are second generation cephalosporins,third generation cephalosporins,first generation cephalosporins,fluoroquinolones and penicillin/beta lactamase inhibitor combination, accounted for 29.69％,22.57％,20.33％,4.66％ and 4.47％ of total DDDs of antibacterial drugs.Individually, the top 10 antibiotics are cefuroxime (12.21％), cefazolin (8.31％), ceftriaxone (7.74％), cefathiamidine (7.34％), cefotiam (4.87％), ceftazidime (3.68％), amoxicillin/clavulanic acid (3.63％), levofloxacin (3.59％), cefoxitin (3.56％), flucloxacillin (3.52％);gentamicin (2.27％), ornidazole (2.00％) and cefoperazone/tazobactam (1.44％) were used most in their categories respectively.The variety and quantity of antibacterial drugs used for different trauma patients were different.Conclusion The trauma score based on ICD-AIS can reflect the severity of trauma.The use of antibiotics in patients with different trauma score can provide reference for the clinical applications of antibiotics in wound care.
RÉSUMÉ
Based on current data security problems existing in medical big data platform,the paper starts from security situation first,analyzing the key points,difficulties and common protection measures of security guarantee of domestic medical big data.Combining with the medical big data platform practice implemented by Southwest Hospital,it puts forward security strategy for data security of current medical big data platform,including management security,decryption security,storage security,network security,etc.
RÉSUMÉ
Objective To analyze the NIH data sharing repository in order to provide reference for the related stud-ies in biomedical data field and for the development of medical data sharing repository in China. Methods Ten typi-cal data sharing repositories ( such as UniProt, Protein Data Bank, and GenBank) were compared. Their data ac-cess methods, data management and sharing model, and service forms were summarized. Results The data manage-ment links and their process standards were designed according to their characteristics. Conclusion Data service tools can be designed, semi-artificial and semi-automatic data check can be carried out, detail meta-data can be collected, and citation rules confirming to the characteristics of data sharing repository can be established in our country by learning the experiences of NIH in developing its data sharing repository.
RÉSUMÉ
Medical big data, an important strategic source of basic data for a country, will be applied in precision clinical diagnosis and treatment, decision-making support, disease monitoring, early warning and management, and public health service. The application of medical big data technology in our country is to be improved at pres-ent. How to realize the smooth transition of traditional medical data to a big data system and the added value of data by data mining and analyzing is an important problem needing to be solved immediately. The key functions, inclu-ding the general frame work and data center frame work of medical big data application information system, were thus planned and designed in this paper by constructing the regional application technology and engineering labora-tory for medical big data.
RÉSUMÉ
Objective To study the hotspots and frontiers in research of medical big data using CiteSpace.Methods The papers on medical big data covered in Web of Science were analyzed using bibliometrics in combination with CiteSpace.Results The achievements in research of medical big data have rapidly increased since 2012.The nodes in research of big data, system, care, medicine and personalized medicine were bigger.The burst value in research of quality, children and mapreduce was higher.Conclusion The number of current researches in medical big data is greater in medical big data technology, precision medicine, medical and health big data management, medical big data privacy while mapreduce, ontology and hadoop are the frontiers and future trend in research of medical big data.
RÉSUMÉ
The use of internet medical big data is conducive to improving the efficiency and quality of health care services,but also contains the problems about the risk of the infringement of the personal information right and privacy right,the hidden dangers of data security,and how to allocate the economic interests of big data.In the internet medical big data ethics,it should establish the regulation principles from four levels of privacy,confidentiality,transparency and identity.In the legal regulation,it should clear the distribution of the economic interests of big data,define the personal information right and data right and have classification protection.