Diagnosis-assisted medical level to improve CDSS future

TextsChen BeibeiCDSS (Clinical Decision Support System), a clinical decision support system, generally refers to any computer system that can support clinical decision-making, the system fully utilizes available, appropriate computer technology, for semi-structured or unstructured medical problems, through human-computer interaction to improve and improve decision-making efficiencyin recent years, with the development of artificial intelligence technology and its application in the medical field, the clinical decision-making support system based on artificial intelligence appears and gradually lands in major hospitals and primary medical institutions, assisting doctors to make clinical decisions, which has received wide attention from the industryThis paper will start from the SYSTEM construction and work logic of CDSS, push factors, application and landing situation, facing the difficulties of the four aspects, with a view to the development of CDSS in China's CDSS a systematic combing and a comprehensive presentation, and on this basis, the future development direction of CDSS in China to forecast and look forward totraditional CDSS is based on clinical guidelines, medical literature and other objective data of the knowledge base, based on some logical associations for induction, deduction, reasoning, matching, in order to achieve diagnosis, treatment and other decision-making supportIn recent years, CDSS systems have gradually developed in the direction of a combination of knowledge base and algorithms: on the basis of traditional knowledge library, using artificial intelligence technologies such as machine learning and big data mining, we can independently acquire knowledge from historical experience and constantly updated electronic medical records data, identify and learn certain models, and thus provide decision-making supportThe system construction of theCDSS consists of three modules: the first is a database, including the medical knowledge base and patient dataFirst, it is necessary to build a medical knowledge base by obtaining a large amount of literature and clinical practice evidence, and the knowledge base must be updated and maintained with the latest development of medicine; Without medical knowledge base and patient data, assisted diagnosis and treatment cannot be started, so data resources are the core of CDSS system construction and operationThe second module is the medical knowledge mapIn layman's terms, it is through the deep learning of the machine that allows AI to understand the medical logic, like a doctor, to understand the whole process of a disease from doubt to diagnosis to treatment, and how to make judgments based on existing data and knowledge in the processThe third module is prediction and demonstration, also known as human-machine communication interface, after mastering the patient's clinical data and disease diagnosis and treatment logic, the prediction results in the appropriate form feedback to the doctorthrough the above three aspects of the construction, CDSS can reside in the cloud server, embedded in the hospital EMR by way of the web, in the doctor's operation of EMR, can provide medical knowledge base retrieval, treatment plan recommendation, similar medical records recommendation, auxiliary diagnosis, medical advice quality control, clinical early warning and other functions, from pre-diagnosis to diagnosis to doctor to provide continuous support, improve the efficiency of the doctor's diagnosis and diagnosis, reduce the rate of misdiagnosisHigh misdiagnosis rate, repeated diagnosis and treatment, uneven distribution of medical resources is the cause of the people "difficult to see a doctor, expensive to see a doctor" is an important reason, but also "healthy China" goal to face one of the challengesSince 2015, the state has issued relevant policies on medical informatization, intelligent medical care, intelligent hospitals and so on, and has put forward the relevant requirements for improving the level of medical informatization and improving the capacity and quality of medical services through various emerging information technologies such as artificial intelligence and big dataIn addition, the state has given strong support to primary health care in recent years, and has issued relevant policies such as graded diagnosis and treatment, medical associations and general practitioner training, with a view to improving the level of diagnosis and treatment of primary medical institutions and assisting in the deep transformation of primary medical institutionsCDSS, as a product closely related to artificial intelligence and medical informationtechnology, has the practical function of assisting primary doctors to make diagnostic decisions and improve their diagnosis and treatment level, which will have broad prospects for development and market opportunities driven by national policies, electronic medical records, as a new form of medical records in the information age, are important indicators to measure the level of modern hospital management and hospital informationizationThe National Health and Care Commission issued in January 2018 the "Further Improvement of Medical Services Action Plan (2018-2020) Assessment Indicators" and august 2018 issued "On the further promotion of electronic medical records as the core of the information construction of medical institutions" requirements, in the electronic medical records information construction work, clinical path, clinical diagnosis and treatment guidelines, technical norms and drug guidelines and other authoritative clinical knowledge embedded in the information system, improve the level of clinical standardizationIn December 2018, in the notice of the "Electronic Medical Record System Application Level Grading Evaluation Management Measures (Trial) and Evaluation Criteria (Trial)" issued by the National Health and Health Commission, hospitals with electronic medical record classification evaluation level severity require clinical decision-making support functions, which indicate that medical decision-making support has become an important link in the construction and rating of electronic medical records in hospitals, laying a solid foundation for the development and landing of CDSSProblems generate demand, and demand stimulates policy-makingChina's uneven allocation of medical resources is very prominent, according to the 2019 National Health Statistics Yearbook data can be seen, public hospitals, accounting for 18.8% of the tertiary hospitals bear 60.8% of the number of patients, the average of each three-level hospital to bear 820,000 visits per year, which leads to the tertiary hospital doctors overburdened, too much pressure, and CDSS applied to the hospital, can assist doctors to make clinical decisions, and improve clinical efficiencyat this stage, there are 997,000 medical and health institutions in China, of which 944,000 are primary health care institutionsPrimary health care institutions account for about 95 per cent, but they have only 32 per cent of the country's health servicesIn developed countries, primary clinics can deal with 85%-90% of patients' health problems, only 53% of the domestic medical treatment by the primary medical institutions, compared with developed countries, China's primary medical institutions more than the number of medical treatment capacity is low, the literacy of doctors in primary health care institutions is low2019 China Health Statistics Yearbook data show that the proportion of community health service centerundergraduates and above doctors accounted for less than half, only 20.9% of township doctors for undergraduate and above, college and secondary education accounted for up to 76.3%, village health room undergraduate and above education of doctors accounted for only 3.1%, college and secondary education of doctors accounted for up to 94.2%This is also a major reason to limit the ability of primary doctors to diagnose and treat, leading to a high rate of misdiagnosis, patients do not want to choose a major reasonunder the strong support of the national policy for primary health care, the basic medical staff need sewage to improve their ability, and the improvement of primary medical care is an important way to alleviate the current disparity in medical resources and improve the current medical situationThe goal of CDSS is to help doctors better diagnose and treat, and this differentiated level of doctors gives CDSS a better place to use in Chinathe ability of primary doctors is not enough to undertake the needs of policy, so how to improve the level of primary-level doctors' medical treatment is a key consideration in the current primary health care construction, the introduction of CDSS is an important development directionscientific clinical decision-making requires clinicians to have multidisciplinary, multi-disciplinary medical knowledge, and in reality, most doctors lack comprehensive clinical dialectical thinking ability, professional level is not high, in diagnosis, often only from their own professional start, rarely take into account other professions, resulting in missed diagnosis, misdiagnosis, this phenomenon is particularly evident in the primary medical institutions But at the moment, there is a serious shortage of GPs in our country, a shortfall of nearly 400,000, and it takes about five to 10 years to train a GP, so many specialists in primary care need to be converted to GPs because of policy requirements If the decision support function is introduced in the primary health information system, through CDSS to assist the doctor's diagnosis and treatment decision-making, we can make up for the lack of professional level of doctors to a certain extent, improve their level of diagnosis and treatment, so as to speed up the progress of training and shorten the cycle of doctor training In addition, CDSS can help the grass-roots to establish homogenized, standardized medical pathways, help grass-roots doctors to avoid some misdiagnosis and missed diagnosis of common diseases, and help them to carry out scientific referrals, help improve the quality of primary health care services, better promote the implementation of graded diagnosis and treatment policies, and alleviate the uneven allocation of medical resources to better serve the grass-roots doctors, improve their level of diagnosis and treatment, which is also the greatest value of CDSS in China Hospitals and primary medical institutions are the main landing places for CDSS enterprises From a business point of view, the development of a specialist version of the CDSS system service hospital and the development of a general version of CDSS system service primary medical institutions is currently the two major development directions of CDSS enterprises From the application of products, the current CDSS primary procurement object or large hospitals, specialists are the core user groups With the implementation of national graded medical treatment and primary health care-related policies, enterprises that provide CDSS services for both specialists and primary-level doctors (such as Hui-Per-Medicare and Baidu Lingmedicine), as well as enterprises that provide CDSS services specifically for primary health care institutions (such as Shenzhen Evidence-based Medicine), are increasingly emerging It can be predicted that with the state's financial and policy support for primary medical institutions continue to increase, CDSS's broader market will be at the grass-roots level CDSS as a new thing just emerged in recent years, most hospitals and primary medical institutions to their understanding of the degree is not enough, the market penetration rate is still very low, there is no unicorn enterprise, and AI diagnostic capacity is still in the continuous verification and optimization, the current state, for enterprises, the primary and fundamental point of competition or products, for market segments, targeting user needs, so that the research and development of products in the diagnostic capacity has a significant advantage, can provide doctors with real help Secondly, each enterprise can combine their own advantages for promotion and commercialization of products, traditional medical information enterprises can use their customer base for product promotion and cooperation, BAT enterprises can use their own advantages in technology, capital, brand and other aspects to increase product research and development and promotion although CDSS has been developed to a certain extent, it still faces many challenges and difficulties, including the following aspects the premise that CDSS provides value is that it must significantly improve the process or results of clinical work, how to obtain and present the clinical application effect of CDSS is a difficult yon Because different CDSSs are designed for different purposes, there is no one-purpose evaluation criteria, which creates difficulties in evaluating the value of the system The willingness of users to spend significant amounts of money to build CDSS can be affected when it is difficult to prove that CDSS can be effective in improving clinical workflow or results The construction of the knowledge base is the core of CDSS, its knowledge acquisition is generally derived from the medical literature and clinical practice, the knowledge involved is very broad, basic medical knowledge, clinical guidelines, evidence-based medical evidence, medical dictionary, medical map, computational tools and other large amounts of data are indispensable, it is not easy to build such a complete, authoritative, integrated multi-field knowledge base Second, medical knowledge is growing, new drugs and diagnoses are being discovered, evidence-based guidance changes with the accumulation of new evidence, and integrating frequently updated knowledge into existing databases is also a challenge CDSS on the data requirements are very high, the successful CDSS should be seamless integration with the existing clinical workflow, patient information system, in order to make decisions at the time and place of automatic advice, in this process, the understanding of the semantic level of different systems is necessary, and the current level of medical information technology in China is uneven, the information system of each hospital is not the same, data standards, structure is not uniform, the lack of medical semantic senforcement standards lead to the implementation of CDSS, the cost of the current use AI software to support decisions is accompanied by legal liability issues: Who is responsible if the decision support system gives inappropriate advice that causes problems for patients? Is the software designer, the provider of medical knowledge, or the provider of the medical service responsible for the final clinical decision?? At present, the laws and regulations in the field of artificial intelligence are not perfect, the division of responsibilities is not clear, how to better regulate and regulate CDSS, and related legal liability division will remain a long-term proposition in this field the future, to better cope with and solve these problems, the need for government, enterprises, medical institutions to work together The government needs to improve the relevant laws and regulations, evaluation system, CDSS suppliers need to effectively meet the needs of end-users, build a sustainable development system, medical service providers need to improve the operating level of CDSS applications to ensure that CDSS is used effectively All in all, the development of CDSS industry is still in a stone-crossing stage, challenges and opportunities coexist, the future has a long way to go, its real performance and impact is worth looking forward to at present, the "2020 Smart Hospital Development Status and Trends Report" and "China's Primary Health Care Research Report 2020" under the leadership of Euda Health are under way