Internet Healthcare (vi): How to open the Chi market for large data?

In Beijing 301 Hospital Computer Application and Management Section of the Deputy Director Zhakingli's desk, there is a pile of "Data query statistics application Form", there is to inquire "environmental factors on respiratory disease" related to the medical records, there are to be transferred allogeneic organ transplantation of the past cases, various inquiries need a wide variety.

"There are more and more applications, and each request is different. To fetch the relevant data in our database, I must write a separate retrieval procedure for each application. "Zhakingli told NetEase Technology," feeling by hand processing more and more coping. Six months ago can return the results two or three days, now the application volume soared, I was part-time, work too busy, those conditions are complex to wait one months to get results. ”

The problem facing Beijing 301 Hospital is a microcosm of the general problems faced in the utilization of medical data in China: On the one hand, the medical informatization process since the 90 's, make more and more medical data accumulate in the information system; On the other hand, for how to exploit and use the data, But there are no mature tools and methods. With good Doctor group chairman Gao Zhan words to generalize, is: "Big data industry is alive, and fire extremely." But the big data industry's large data production is dead. ”

At present, the medical large data has shown the very broad application foreground to the human. Traditional medicine is a small data judgment and decision-making, the correct rate is difficult to guarantee, completely rely on the doctor's experience and ability. Previous diagnoses based on statistical significance will be replaced by the diagnosis of large data based on personalization.

But in the past year, Gao Zhan found that 50%-60% of the English language is already exploring "how" to bring value to large data, and there are many useful conclusions, such as "Caesarean section is the best time before and after the due date of two days", but the Chinese information is still in preach big data "can" produce value this level.

The future market is undoubtedly huge, Shanghai Jin Shi Tatsu Ning Software Co., Ltd. Deputy general manager Xu Chunhua estimated that the future medical health data Analysis Services market will reach Chi.

But the question is, how can this market be opened?

Status quo: Capital is being laid

The scramble for future capital for big data has begun. According to media reports, from 2013 to 2014, large data is one of the main areas of Internet health and investment.

Formerly known as China's first professional hospital information system software development and application engineering enterprises of the Peking University Medical Information Technology Co., Ltd. (hereinafter referred to as "Beijing University Medical Letter") has been targeted at medical large data strategic direction.

In March this year, the Beijing University Foundation invested 30 million in the Beijing University medical letter, the biggest investment in the history of Peking University. On March 14, the founding meeting of the Medical Letter of Peking University, founder Group senior Vice President, founder Letter Production Group CEO of China directly to its entrusted with the task: "Beida and founder Group of Strong Support, the great data era brings unlimited opportunities, should let us feel that our cause is no glory; We must work together to make the mission of Beida and founder to the medical letter of Peking University complete well and do great data business, not only to be a big data center of Beida, but also to be a national and world-class data center in the future. ”

December 11, Hong Yi Investment, Goldman Sachs, Neusoft Holdings and collaborative innovation and other investors jointly on the east soft city of 170 million U.S. dollars in capital increase. Neusoft is a company that specializes in Internet medical and health management in East Soft group dedicated to providing a health management and medical service platform based on O2O mode through large data, cloud computing, IoT, and mobile Internet, this investment refreshes the record of the largest single financing in the domestic Internet medical and health management field, is also the world's largest internet health and health management one of the biggest single fund.

The Shanghai Medical Union Project has established the current clinical information sharing system with the largest network of medical institutions. The construction of the contractor Wanda Information Co., Ltd. (hereinafter referred to as "Wanda Information"), July 2014 acquisition of Shanghai Forte Computer Technology Co., Ltd., August acquisition of Ningbo Golden Tang Software Co., Ltd. Both companies have been in the field of medical information technology for more than more than 10 years of business.

Wanda Information Co., Ltd. President assistant Feng Dongrei told NetEase Technology, Wanda Information plus the new acquisition of two subsidiaries, now a total of 3500 employees, of which engaged in health and health related to the Division and subsidiaries, the staff has 1500 people, is the largest information business in the biggest piece.

Shanghai Jin Shi Tatsu Ning Software Co., Ltd. is a professional engaged in the field of medical and health information, digital, software research and development of High-tech enterprises. Xu Chunhua told NetEase Technology, Jin Shi Tatsu ning Do data processing is an extension of the original business, but over the past year, he has found that not only traditional medical it hardware and software services, but also many cross-border, former and health-care listed companies involved in the field of medical data, such as the real estate business , concentrated juice production enterprises, such as investment in China.

In the "2014 China Mobile Medical Industry Annual Meeting", China Mobile, China Telecom, Unicom, IBM, insurance companies, and other health management companies are involved in the theme of "Health Large Data Health" forum.

Outside the enterprise, hospitals and various research institutes are also trying to carry out research on large medical data. November 29, the Institute of Advanced Technology of Shenzhen Academy of Sciences, the health of large Data research center was established. Beijing University is preparing to set up a large medical and health data center in Beijing. In the last two months, Feng Dongrei has visited universities such as Beijing University, Zhejiang University, Sun Yat-sen University and South Central College, "These colleges and universities want to work with us on big data." "He revealed.

October 18, 2014, the Capital Medical University affiliated Beijing Anzhen Hospital and Pfizer Investment Co., Ltd., the first domestic major cardiovascular medical data center project started.

Clinical application: Immature

The current demand for large medical data is concentrated at three levels: operational management, adjuvant therapy, and ancillary research. In the industry's view, at present in China, the medical data has achieved good results is the administration.

Beijing Public Health Information Center Statistics Room director Guo Mening told NetEase Technology, the current data mining and utilization, Beijing Public Health Information Center to do more effective is to the medical institutions performance analysis.

In the past, it was not easy to evaluate the performance of the medical institutions because each patient was in different conditions, the difficulty of working in the medical institutions and the effectiveness of the work were difficult to measure, and the contrast between medical institutions was very difficult. Guo Mening told NetEase Technology, in the past, the usual mode of performance evaluation for medical institutions is to find experts to the hospital score, based on experience and subjective judgment more.

Starting from 2008, the Beijing Public Health Information Center attempted to evaluate the performance of medical institutions in the area according to the indicators obtained through data mining. She told NetEase technology: "The use of statistical methods, medical institutions can treat patients with similar illnesses grouped, so that in the same group of patients, you can compare the quality of the services of various institutions." This can promote the meticulous management of medical institutions and improve the quality of medical services. ”

Guo Mening told NetEase Technology, the future in the public health field, medical data mining and utilization of the prospect is very broad. For example, in Health resources planning, allocation, disease early warning and so on will be fully applied.

There are already institutions in the field of clinical adjuvant therapy and assisted research that are exploring and experimenting, but are not yet mature.

Wanda Information developed before the "Clinical decision system" in the industry more well-known, at present, 38 municipal hospitals in Shanghai can provide doctors with recent repeated medication, inspection, inspection reminders, treatment safety warnings and clinical pathways (refers to the establishment of a standardized treatment for a disease model and treatment procedures, is a comprehensive model of clinical treatment, guided by evidence-based medical evidence and guidelines to promote the treatment of tissue and disease management methods, and ultimately to standardize medical behavior, reduce variability, reduce costs, improve the role of quality services.

But, Feng Dongrei told NetEase Technology, the provision of these functions, in fact, need to do the building of knowledge base, but now the knowledge base is based on existing textbooks, drug use instructions and some clinical manuals, are relatively simple, in the future to the doctor's real experience in the hospital input, in order to play a greater clinical significance, This is the real big data mining. He revealed that cardiovascular and oncology experts were involved in the upcoming project "Research on the analysis and application of major cardiovascular disease and cancer disease data".

2012, Wanda Information, the National Health Engineering Center applied for the Shanghai Science and Technology Commission medical and health large data topic. In this project, Wanda Information attempted a large data analysis of hypertension in an attempt to find a link between illness, medication and efficacy. But Feng Dongrei told NetEase technology, such analysis is now encountering some difficulties. Previous clinical studies (RCTs of randomized controlled trials) were performed in the experimental and control groups, while the control group was a relatively desirable control population excluding complications. The sample size is small, but the data particles in each sample are very thin. But using large medical data for analysis, the sample size is very large, but each sample data particles are thicker. Therefore, the research methods of traditional RCT cannot be applied. Therefore, new research ideas are needed. At present, with large data has found some therapeutic means and the relevance of the effect, but this conclusion on the clinical use of how much significance remains to be tested.

Such research continues. The national "863" Plan 2015 Annual Project Declaration Guide, in the field of biological and pharmaceutical technology has been deployed "large biological data development and use of key technology research", covering the content of biological large data standardization and integration, fusion technology, large data presentation index, search and storage access technology, Analysis and application of large data processing of cardiovascular disease and tumor disease, based on regional medical and health data processing analysis and application research, the group of large datacenter and knowledge base construction and service technology.

Wanda Information participated in two of them, "based on regional medical and health data processing analysis and application research", as well as "cardiovascular disease and cancer disease large data processing analysis and application research." The latter carried out a study of large data analysis and application of cardiovascular disease and tumor diseases facing TCM.

Peking University Medical Letter is also cooperating with Peking University to study the analysis and utilization of clinical medical data. There are more than 500 hospitals in Peking University Medical Service, among them, more than 200 hospitals, accounting for the total number of three hospitals in the country about 1/4, Beijing University affiliated with 9 affiliated hospitals, 13 teaching hospitals, these hospital information systems accumulated a large number of data for the analysis and use of large data has laid a solid foundation.

Peking University Medical Letter senior Vice president and CTO Shang told NetEase Science and technology, at present the clinical decision support system of Peking University medical Letter is in Peking University People's Hospital, Peking University International Hospital, Jiangsu Province People's Hospital carries on the pilot.

The University medical letter has developed clinical early warning and the application of the recommendation class. The application of early warning can judge the risk of the disease and prompt it according to some vital signs of the patient. Recommended class of application, the current University of medical letters to do diabetes This disease, the system can be based on the symptoms of diabetes, test results and medical records, give the corresponding treatment plan recommendations.

In the field of clinical research, the University medical letter has also done some data analysis, and obtained some results. For example, the previous misdiagnosis rate of endometriosis and fibroids was as high as 65.1%, because the symptoms of the two diseases were very similar. Through large data analysis found that ovarian cyst, abdominal pain, anemia in the three symptoms of these two diseases in the weight is different, endometriosis and ovarian cyst is the strongest association, uterine fibroids and anemia are the most relevant.

"We have analyzed the results, but how to use them clinically, we need to explore further." "Shang introduced.

Good Doctor group chairman Gao Zhan believes that in order to make large data produce value, need to have a complete value chain, the current China's value chain is still missing. The value chain of large data has four links of data collection, storage, analysis and application, but at present the industry input is more collection and storage, and the analysis and application are weak. The quality of the data is not very high, even in the process of data collection, due to the lack of corresponding mechanisms.

The industry said: how to do

In the current situation, how to do the big data? Gao Zhan believes that we should first grasp some of the key business needs, while the data base is better, do it first, and then gradually expand. He cited two examples.

Good Doctor Group has participated in a project of Anhui province Feixi County Health Bureau. According to Gao Zhan Introduction, Feixi County Health Bureau of the new Agricultural and medical reimbursement system data and Health Bureau for the residents to establish an electronic health file to do a correlation analysis. The results showed that the residents ' obesity and smoking were correlated with hypertension and diabetes. Gao Zhan told NetEase Technology, this is not a new discovery, but the significance of applying large data analysis is that previously people were only conceptually aware of the effects of obesity and smoking on high blood pressure and diabetes, and after using large data analysis, it was true to see that a specific individual's obesity and smoking had an impact on the disease. After that, the Community hospital applied this analysis, to patients with hypertension, diabetes patients not only lower blood pressure, reduce blood sugar, but also to intervene in patients with obesity and smoking. Now the whole Feixi, as well as many counties in Anhui are promoting such a practice, this a small change, so that the use of electronic health files in many places from around 20% to 60%-70%, doctors do not use the active database was activated.

Feixi also made an experiment. The original rural medical treatment, regardless of adult children, are generally open antibiotics, infusion, good doctors group for many years to do village doctors training, call for the abuse of antibiotics, but in practice, not very effective. Earlier this year, Anhui Province launched a primary health care institutions prescription set system, the system can be based on disease diagnosis, suggest prescriptions, to standardize the treatment and medication behavior. Meanwhile, since February this year, the Feixi County Health Bureau has issued a monthly proportion of prescriptions for doctors prescribing more than two antibiotics. As a result, in October, up to 20%-60% of the data were dropped to single-digit digits.

Gao Zhan Summary, large data application should first from "low hanging fruits", that is, hang the fruit, easy to achieve the goal to start, the limited "small data" to use well.

This view coincides with the original director Ninlian of the computer room of Beijing 301 Hospital.

Ninlian told NetEase technology, the current medium-scale hospitals have accumulated at least hundreds of GB of data, each 100GB of data is equivalent to 300,000 medical records. Although this order of magnitude has not yet reached the PB level, it must contain a lot of valuable information.

He took his observations for example: "I observed 20 people around me with malignant tumours, and I found that six or seven of them have had a very distressed period of life." 20 samples, you can sum up a little rule, not to mention this sample size to GB? That will result in more, more valuable, and more accurate conclusions. ”

He advocates the use of the current data, and now 301 of the hospital's pile of applications is the real demand. He said: "At present in the hospital, this kind of service is still passive, is the doctor to come to us to provide the service, if this kind of service further takes the initiative, advertise, tells the general health care personnel and the management personnel we can provide you to need Information services, doctors and technicians in the process of continuous communication, will be able to excavate large data greater value. ”

Ninlian is 80 years old, but the word pops up in his mouth is "quick iteration," according to his statement: "Good tools are used." This is a huge market and we can't wait. ”

PostScript

Since the second half of 2013, Internet health, especially in the field of mobile medicine, has indeed been a fire. The mobile internet is like a ribbon that links all aspects of the medical process to patients.

The main effect of this ribbon at present is to improve efficiency. For example, we can not go to the hospital to hang the number, in the handheld hospital can consult doctors, do not go to pharmacies can also buy drugs online, or even lower prices.

Such a change has not yet touched the heart of medical activity. For example, although the registration platform is everywhere, but the number of the best doctors in these platforms is still difficult to find traces, in the handheld hospital can only do simple disease counseling, more accurate diagnosis still depends on the hospital, because of prescriptions, Medicare reimbursement and security issues, many people do not go to the online pharmacy to buy medicine.

The subversion of medical care by the Internet depends on the openness of the policy and the speed of the technology, which are mutually promoted.

But we are also delighted to see that so many people are working hard for this "subversion", who are coming into our stories of entrepreneurs who are eager to change outside the hospital and doctors and managers who are pushing for change inside hospitals.

The possibilities offered by Internet medicine are also extremely tempting, we may be able to use this tool to record the life of all the health data, these data will reveal the law, create value, we can under the guidance of doctors and technology, the full and continuous management of their health, do not wait until the disease to go to the hospital, Even in the future, we can predict our susceptibility by genetic testing, take measures early and avoid getting sick, and now the technology for gene testing is the price we can afford.

The important thing is that this future may not be so far away, we can see a lot of real changes in our lifetimes, even in the coming years, we can expect rapid changes.

The two-month interview ended. We are trying to sketch a picture of Internet medical care in China in the present time, only to find that many times it is exhaustive and difficult to Biching. We, like everyone else, are looking forward to changing our current medical pain points, and we will continue to focus on this process in the future.

(Responsible editor: Mengyishan)