How to implement the monitoring intensive care process for the ICU to turn on the red light
When the patient enters the intensive care unit, it is equivalent to entrusting everything to the doctor. However, there are many indicators for monitoring in intensive care units, and the weights of indicators that different patients need to pay attention to are different. Although doctors and nurses concentrate on handling patient-related matters, mistakes are still inevitable. According to the data, the incidence of ICU errors in the United States in 2005 was as high as 1497 times per 10,000 people, of which 13% were seriously life-threatening; 1.7 medical errors per person per day, and an average of 98,000 ICU patients died each year. Medical errors. However, there is also a positive data showing that 28% to 84% of these medical errors are preventable, but the hospital needs to find an effective way to prevent it. The domestic situation is also not optimistic. With the advent of aging, the number of patients is increasing, and doctors are growing for a long time. The gap between supply and demand is getting bigger and bigger, and the resources of professional doctors are seriously insufficient. On the other hand, many diseases and treatments currently lack professional guides for the domestic population, which are judged by doctors' experience, and experienced clinicians are very limited. The situation of the ICU is even more severe. For doctors, the patients entering the ICU are generally complicated and complicated. The comprehensive data dimension is as many as 236, far exceeding the scope of human control, and it is difficult to accurately assess the changes in the condition. For the patients, the cost of the ICU is high, and the per capita consumption of medical expenses is 75,673 yuan (the annual medical expenses are more than 100 billion yuan). It can be said that the ICU is the most "burning" department in the hospital. In view of the pain point of the current hospital ICU, the medical intelligence town located in Hangzhou artificial intelligence town built a disease assessment system based on ICU medical big data , using artificial intelligence to help predict the patient's disease development trend, and derived the ventilator intelligent management system, Acute kidney injury (AKI) early warning system is used to provide doctors with auxiliary warning and decision-making to improve the survival rate and prognosis of patients. At present, the development model of Maixing Medical has empowered the AI ​​of the ICU industry through the accumulation of data structure, and strives to become the "zero-tech" of the ICU industry. Focus on the clinical diagnosis system with ICU as the core Pulse Xing Medical focuses on building a multi-parameter intelligent medical big data platform to provide data sources for medical big data platforms and AI product development; and using clinical data to learn through NLP (natural language processing), image recognition, predictive modeling, etc. The method is to develop a clinically available medical aided decision-making system to assist doctors in accurately determining the patient's condition. 1. Disease assessment and death prediction system At the core of the core product assessment system of Pulse Medical is a disease assessment model based on the MIMIC database. It can analyze clinical data in real time, display the parameters most relevant to the disease to doctors according to the weight, and predict the trend of disease development and death risk. Auxiliary doctors use their clinical experience to quickly make clinical decisions. The death prediction model developed by Pulse Xing has a prediction accuracy of 94% for the patient's death results, and the accuracy rate has reached 87% when the example is verified by the Union Hospital. 2, WeanDoc (artificial intelligence to build a ventilator intelligent management system) In the ICU department, more than 50% of critically ill patients may use mechanical ventilation, and the parameter adjustment of the mechanical ventilation ventilator and the timing of the withdrawal timing are more demanding for the clinician's experience. WeanDoc can select the appropriate mode at the beginning of the ventilator use, set parameters to provide reasonable respiratory support to maintain and improve the patient's breathing and physiology, and prolong the valuable treatment time for the treatment of primary disease. Different regimens were used to set tidal volume (VT), positive end expiratory pressure (PEEP), inhaled oxygen concentration (FiO2), and respiratory rate (RR) for different patients. When the patient's condition is improved, the doctor is advised to reduce the respiratory support properly when appropriate, so that the patient can gradually resume spontaneous breathing, shorten the use time, reduce the dependence on the ventilator, and improve the prognosis of the disease. All the indicators monitored by WeanDoc are continuously monitored by the system, and the risk of infection, patient suffering, medical expenses, and ventilator use time are significantly reduced, and doctors can reduce errors even with the help of WeanDoc, even It is a mistake in the direction of parameter setting, and it can correct the error immediately and avoid serious impact. At the same time, this means that the respiratory therapist with less medical experience has obtained a 24-hour companion teacher, and the talent gap of the hospital will be reduced accordingly. Partner is one of the key factors for success In order to understand the actual situation of the ICU early warning products of the pulse medical treatment, the arterial network reporter came to the Internal Medicine ICU (MICU) of the Union Hospital. The project is led by Prof. Du Bin, the director of the ICU of the Union Hospital, and led by Prof. Du Bin. The President of the Association of Disease Medicine (APACCM) and the President of the Chinese Medical Association's Critical Care Medicine Branch have a very high status in the ICU industry. As a partner of Maixing Medical, Weng, the deputy director of the ICU of the Union Hospital, talked about the pain points of the database today: "In terms of data, each hospital in Beijing has its own huge database, but many of the databases are not perfect. Non-standard, it is difficult to export for clinical research, so the first thing we have to do is to make a standard database, but it is not enough to talk about standardization. Someone needs to test the water first." Pulse Xing Medical and Xiehe have jointly established the ICU scientific research big data platform, which enables ICU doctors to realize the independent use of ICU clinical data and conduct relevant scientific research. At present, the research platform of Maixing has been deployed in several hospitals in Beijing. The demo version was officially launched on August 31 in the case show of the official website of Pulse Xing. It is hoped that more ICU departments will join in to participate in the formulation and maintenance of standards. Associate Professor Weng Li will give a special report entitled "ICU Specialist Database Establishment" at the China Critical Care Medicine Conference on September 8 this year. Concord Hospital pays great attention to the standardization of data in ICU. Its data structure is complete and standardized, and it can be easily exported for algorithm learning. It is very expensive for enterprises to spend time processing data, because some data is due to some indicators. The lack of it, even with the heart repair can not get the effect of the target. “The MICU of Concord Hospital can provide not many cases, and the total data of more than 2000 patients. However, the ICU data is different from the regular medical data. With the heart rate as an example, we record the data in 5 seconds, and there are nearly 17280 data in one day. If you look at the entire timeline, the amount of data in each patient's file is very large enough for algorithm learning. In addition, after we train the algorithm, it is not only used for testing our own database, but also for other The hospital performs verification so that we can immediately find the error and adjust the algorithm." Therefore, the cooperation object of Union Hospital is to lay a good foundation for the promotion of medical AI head enterprises. Implantable is the key to entering the hospital There are many companies that focus on medical information , but there are very few AI-assisted tests. However, AI empowerment can't be the reason for hospitals to choose AI companies. Generally, large-scale information companies have built a complete set of information systems including HIS, PACS, RIS, CDSS, etc. Once the hospital chooses this solution, There is no reason to pay for a single AI-enabled CDSS system. Pulse Xing Medical clearly understands the needs and problems of hospital informationization. This integrated research platform + AI clinical application system can be used in clinical applications by accessing the information interface of information technology enterprises through cooperation with medical information enterprises. The core lies in the decentralization of the grassroots Compared with the top three hospitals, the primary hospitals have a greater demand for this ICU early warning system. Restricted by the limitation of resource allocation, grassroots hospitals lack corresponding talents, and corresponding medical accidents are more frequent. On the other hand, the size of hospitals also limits the inflow of medical talents. This is a vicious cycle. Frequent accidents have reduced the prestige of primary hospitals and exacerbated the motivation of residents to go to the top three hospitals. This is actually not conducive to the implementation of the stratified medical treatment policy. The early warning system of Maixing Medical can not only reduce the incidence of accidents in primary hospitals, but also improve the impression of primary care in the hearts of residents. It can also continuously give doctors and nurses the key to making mistakes. Frequent corrections can make doctors and nurses have medical standards. Get a substantial boost. All along, doctors are the core of the hospital. The doctor's every move is related to the hospital's reputation and development. When AI is part of the doctor's work, it can reduce the technological gap between the primary hospital and the top three hospitals to some extent. A long-term advancement. Future The death warning system is only the first step of the pulse medical treatment. More importantly, through continuous machine learning, we can not only provide warnings to doctors, but also give corresponding decisions. Deputy Director Weng Li told the arterial network reporter: "This system can help us correct the error immediately, but what should we do when the error occurs? This is unpredictable. Doctors can only use the relevant parameters of the equipment according to their own experience. Adjustments, but there is no guarantee that this adjustment will have a positive impact on the patient - this is exactly what we want to study. Through the continuous analysis of the database by AI, we may be able to find out the rules and make the most beneficial to the patient. Decision making, but this requires a large database. We also hope that more hospitals and doctors join us to establish industry data standards and promote the intelligent development of the ICU industry." So far there is still no convincing standard to evaluate the database, but we can't wait until the official standard is set and then make a decision. There is no harm in the first step. In the following, Maixing Medical will cooperate more deeply with the hospital, from the perspective of doctors and patients, to help doctors research, improve the interaction design of existing systems, give doctors a better experience, and provide patients with more secure ICU. service. On the other hand, Pulse Health will also help doctors to explore the long-term impact of changes in ICU indicators on patients, and to explore precise solutions for patients with worsening conditions, making ICU more intelligent. Sunflower Kernel,Sunflower Seed Kernels,Flavored Sunflower Kernels,Raw Sunflower Kernels Wuyuan county dafeng oil food co.,ltd , https://www.dafengfood.com.cn