Si Nuo may not have thought that after nearly two hundred years, the application of big data is no longer accidental. With the rapid development of medical and health informatization, through the combination with AI, its penetration in biomedical research and development, disease management, public health and health management has gradually become the norm, but the problems have also become correspondingly prominent.
Information islands still exist.
In the past two years, there have been frequent medical big health data policies, and relevant guidance opinions have been put forward from top-level design, specific planning guidance, data privacy and security, data management and other aspects.
From June 2065438 to June 2006, the General Office of the State Council issued "Guiding Opinions on Promoting and Standardizing the Application Development of Big Data in Health Care", pointing out that all kinds of medical and health institutions should be encouraged to promote the collection and storage of big data in health care, strengthen application support and technical support for operation and maintenance, open up data resource channels, and speed up the construction and improvement of basic databases with residents' electronic health records, electronic medical records and electronic prescriptions as the core.
2065438+September 2008, National Health Commission issued "National Health Big Data Standards, Safety and Service Management Measures (Trial)" to standardize the health big data industry from the perspective of standardized management, development and utilization. The "Measures" put forward guidance from four aspects: medical big data standards, medical big data security, medical big data services and medical big data supervision, directly hitting the pain points in the current medical big data field, and it is of great significance for coordinating the standard management of data and implementing safety responsibilities and standardizing data services and management in the future.
However, even with the support of special policies, it is limited to the macro level. Compared with other mature fields, the laws and regulations in the field of health care big data are still obviously lagging behind, lacking comprehensive, detailed and clear guidelines and rules, which seriously restricts its development. Although many enterprises have been deeply involved in the field of medical big data at this stage, they are still crossing the river by feeling the stones due to the uncertainty of market access and industrial policies, and the market enthusiasm and vitality have not been fully and effectively released.
Liu Lei, a professor at the Institute of Biomedicine, Shanghai Medical College, Fudan University, believes that it is the uncertainty of medical big data policy and the inconsistency of standards that directly lead to difficulties in data exchange and information sharing among systems, resulting in a large number of "information islands". For a simple example, the film taken by the patient in hospital A went to hospital B but didn't recognize it. Doctors in hospital B need to start from scratch if they want to know the patient's information. The patient's examination in A hospital needs to be repeated in B hospital. "At least at this stage, it is difficult to get through the clinical big data resource channel between medical institutions." Liu Lei said.
Similar troubles also occurred in the United States, more than 10 thousand kilometers away. Philip Paynes, director of the Institute of Information Studies at the University of Washington School of Medicine, said in an interview with Yigu that the "isolation" of clinical big data has brought burdens to national medical insurance institutions, patients and hospitals. Realizing the interoperability between big data is a problem that is being solved all over the world.
As well-known researchers in two top universities, Liu Lei and Paynes want to make some efforts and attempts in the field of clinical big data.
Some of their ideas were quickly supported by the school level. 2065438+On July 26-29, 2009, the seminar on applied clinical informatics and data analysis jointly taught by Fudan University Medical College and Washington University Medical College in St. Louis started for the first time.
Liu Lei, professor of the Academy of Biomedical Sciences and director of the Institute of Intelligent Diagnosis of Medical Information and Medical Images of the Big Data Institute of Fudan University, gave lectures.
According to Liu Lei, this seminar received positive response from people in the industry. Among the first students, hospitals, medical enterprises and universities each account for one third. "I just want to analyze and be interested in clinical big data. People in the industry gathered together, and through the efforts of * * *, the effective use of clinical big data can be further promoted. "
Professor Philip Paynes, Director of Institute of Information, Washington University School of Medicine, St. Louis.
"I hope that through this international cooperation, clinical big data can really' run' between medical institutions and even across borders." Paynes said.
Their respective explorations
Before this possibility, Liu Lei and Paynes' respective research institutions have done a lot of work.
It is reported that Liu Lei's biomedical research in Shanghai Medical College of Fudan University is committed to establishing "the first domestic and world-class biomedical interdisciplinary research institution", which has formed three dominant directions of "Molecular Cell Biology of Metabolism and Tumors", "Medical Epigenetics" and "systems biomedicine", and is striving to expand translational medicine research and precision medicine research, including geriatrics, tumors and cardiovascular diseases, birth defects, target structure and activity.
It is also known that the biomedical research of Shanghai Medical College of Fudan University is still applying for the construction project of supercomputing center to support the research of biological big data. "Fudan University has 17 affiliated teaching hospitals including Zhongshan Hospital, Huashan Hospital and Renji Hospital, and some of them are also doing their own clinical big data centers. From the research level, I hope to provide them with some strong support for talent training and technical research. " Liu Lei said.
Paynes Institute of Information at the University of Washington School of Medicine is the center of all big data projects at the University of Washington. "We have the best genome research institute and the most productive and influential basic scientific research enterprise in the world", and we have strong capabilities in medical information technology, but the integration of big data needs to be strengthened. "This has also become the focus of Paynes' work since he became the first director of the Institute of Information Science at the University of Washington.
Since Paynes took office, the hospital has cooperated with its 15 affiliated teaching hospital, laying a full chain of clinical big data from the collection, integration, mining and application of clinical big data.
According to Paynes, the 15 teaching hospital affiliated to the institute is simply a treasure of big data sources. These 15 hospitals, which rank higher among American medical institutions, produce a large amount of clinical data every day. Retrospective study based on these existing clinical data is one of the most basic and important research methods to analyze and study diseases. Through the statistical analysis of these massive clinical data, the analysis results will in turn provide doctors with the whole process of clinical diagnosis and treatment of diseases.
"Our dream is not only to use clinical big data to help patients, but to hope that these clinical big data can penetrate into their lives and work, and even leisure and entertainment. Through the analysis of big data, the probability of their illness can be minimized and people can always stay healthy. " Paynes is looking forward to Medical Valley.
Future development conception
In the work of integrating a large number of clinical data done by Liu Lei, Paynes and their teams, because they have their own powerful teaching hospitals, the source of data is no longer a problem. However, there are two more realistic problems before them. One is to solve the problem of selecting multimodal clinical big data. Clinical big data is a multi-modal data with various sources, including well-structured data, such as laboratory sheets and prescriptions; There are also some semi-structured data, such as hospitalization summary and discharge summary; There are also completely unstructured data, such as medical images; There are also genomic data such as gene sequencing; And time series data, for example, patients will be seen in ICU with various instruments to measure blood pressure, heart rate, pulse and other flow data.
How to select the required data from these different patterns of data? Liu Lei said that what they need is more structured clinical data. In order to get this part of the data, the data will be cleaned through a certain technical platform to screen out high-quality and effective data.
After this problem is solved, there is still a controversy that clinical big data has been unable to avoid-security and privacy issues.
In this regard, Liu Lei said that relying on the existing technology, the clinical big data collected at present can basically be "not discharged", which ensures the security of the data to a certain extent. Paynes also pointed out that the United States has strict laws and regulations on the protection of medical big data. When the patient's key private data, such as name, address, telephone number and ID number, enter the data management, it must be mosaicked and the data must be strongly encrypted. Even if it is leaked, the data cannot be decrypted. All data access must have a set of strict access control (when and who can access what), so as to ensure the security of data.
When the technical problem is no longer a problem, it means that the combination of clinical big data and AI will be more perfect. Therefore, Liu Lei and Paynes hope that the regulatory authorities can evaluate the effectiveness and safety of AI based on big data training in the future, that is, the approval access should be strict and the public's awareness of medical AI should be strengthened. No matter how advanced artificial intelligence is, it has certain limitations. It can never replace a doctor, but can only be a doctor's auxiliary diagnosis.
Although there is still a long way to go, Liu Lei and Paynes are full of confidence in the combination of clinical big data and AI. At least in their existing work plan, the seminar on applied clinical informatics and data analysis can eventually develop into a master's talent training program to train more professionals for clinical big data and artificial intelligence. At the same time, based on the work carried out by the two research institutions at this stage, cross-border integration and unification can be realized one day, all clinical big data can be unified on the same model, and a consortium similar to alliance data can be established, which will facilitate data integration and application.
All the works indicated on this website are from other media, aiming to convey more information, which does not mean that this website agrees with its views and is responsible for its authenticity.