CC BY
39POLITICAL SCIENCES
USE OF BIG DATA IN HEALTHCARE
Makarenko M.
Candidate of Medical Sciences, Director of MM-Dental Clinic, candidate for the degree of Doctor of Science, Taras Shevchenko National University of Kyiv ORCID: 0000-0001-8677-8670
ABSTRACT
The article analyzes the problem of using Big Data in the field of health care. It is argued that the synergy of the digital economy and digital medicine should have no subordinate or causal links, but exist in parallel, their interaction should be aimed at the benefit of each system, and hence the whole society - both directly and indirectly. This indicates the humanitarian and social aspects of the problem, but there is also an economic and technological aspect. One of the important factors of unsatisfactory management in the field of health care in Ukraine is the low efficiency of information management in the health care system, which is due to the lack and isolation of Big Data registers and outdated filling of statistical forms; insufficient and uncoordinated use of modern information and communication technologies. In particular, data management in health facilities and statistical systems is in some cases unmanaged and in others excessively managed, resulting in unbalanced management of medical data, which results in results that are incomparable to the efforts made; The information space of the health care system is fragmentary, there is no operational data on most health care parameters. It has been proven that the health care system of the last century solved the task of restoring human resources, and the newest national one, in the conditions of reform, is called to increase human capital by supporting changes in health attitudes of the whole society and every citizen. To solve this problem, it is necessary, among other things, to make transformational changes in the healthcare sector through the introduction of the latest technologies. Such changes require a holistic strategy and roadmap, as well as functional systems and projects.
Keywords: public administration, healthcare, Big Data, data management in medical institutions.
Formulation of the problem. It is a mistake to believe that the development of the latest technologies will automatically increase the efficiency of the health sector and improve the provision of medical care. In fact, their introduction does not mean that medical services will automatically become available. The use of modern technologies, such as the Internet of Things (IoT), Big Data (BD) or Artificial Intelligence (AI), provided the wrong health care model can only exacerbate inequalities in the provision of health care - by population, territory, nosological units and the like.
The purpose of health care as a system - the same as in any other social system - is self-reproduction, preferably expanded. The main goal of public administration in this sector should be to ensure that health care provides high-quality and affordable medical care with appropriate technological and resource provision [1, p. 27].
The synergy of the digital economy and digital medicine should not have subordinate and not causal relationships, but exist in parallel. However, their interaction should be aimed at the benefit of each system, and hence the whole society - both directly and indirectly. These are the humanitarian and social aspects of the problem, but there is also an economic and technological aspect.
The economy is made up of organization, technology, and resources. With the development of human civilization, it is the human resource that becomes more and more important. For the new technological structure of the digital economy, it is necessary not just a
Human Resource (HR), but human capital. The difference lies in the fact that a resource is something that must be spent and restored, while capital must be preserved and increased [2].
The health care system of the last century solved the problems of restoring human resources, and the newest national one, under the conditions of reform, is designed to increase human capital by supporting changes in the attitude towards health of both the whole society and every citizen. To solve this problem, it is necessary, among other things, to carry out transformational changes in the healthcare sector through the introduction of the latest technologies. Implementing such changes requires a holistic strategy and roadmap, followed by functional systems and projects.
Presentation of the main material of the study with a full justification of scientific results. The World Health Organization, by monitoring the development of health systems in the countries of the world, determines the urgent need for the existence of an official national policy and strategy in the field of health, which must correspond to government needs and priorities, and acts as a key indicator in assessing the public health management system.
It is known that today technologies allow the implementation of the most complex projects, but they must have the appropriate resource support. In healthcare, it is impossible to mechanically transfer solutions, even successful ones, from other areas of economic activity. Attempting to create an environment of exceptional commercial health financing and competition for resources leads to all kinds of competition, but
not competition in the quality of health care delivery [3, p. 11].
Unfortunately, Ukraine still does not have a scientifically based document, legalized at the national level, created by the program-targeted method, which would determine the national policy and strategy for the development of the health sector and reflect the national needs and priorities in this area. This document should contain a real assessment of available resources, define the vision for the future of the health care system, set goals and objectives regardless of political and financial interests. A consequence of the absence of these important documents may be the adoption of non-systemic management decisions, which carry quite serious risks today and in the future and form mistrust of the ongoing medical reform among medical personnel and citizens.
In 2017, the Cabinet of Ministers of Ukraine launched a pilot project, the main purpose of which was to gain experience in changing the algorithms for financing medical care, which involved some domestic research institutions of the National Academy of Medical Sciences of Ukraine - M. Amosov National Institute of Cardiovascular Surgery, Institute of Cardiology named after academician M. Strazhesk, A. Romodanov Institute of Neurosurgery, O. Shalimov National Institute of Surgery and Transplantology [4].
The results of the implementation of this project can become a very useful experience and prevent managerial mistakes at the stages of modernization of secondary and tertiary medical care in all regions of Ukraine.
Pilot testing of new funding mechanisms for institutions indicates the need to use modern digital technologies related to improving accounting, analysis and implementation of communications in organizing the provision of statutory activities of medical institutions operating on the basis of an enterprise.
At the same time, it is a matter of concern that in the conditions of free choice of medical information systems by health institutions to work with eHealth, the Ministry of Health faces the question of urgent unification of the activities of such systems to ensure their integration with each other, since now the use of data between different systems is impossible even within the hospital district, and even more so in the institutions of the National Academy of Medical Sciences of Ukraine.
In addition, the electronic health care system (eHealth) was launched without creating a system of information protection of the central database, as provided by the Resolution of the Cabinet of Ministers of Ukraine from 25.04.2018 № 411 "Some issues of the electronic health care system", which approved the Procedure electronic health care system [4].
The new regulations of the Cabinet of Ministers provide for full responsibility for ensuring the protection of patients' personal data on service providers. This situation creates significant risks today and in the future and creates distrust of the reform among medical staff and citizens.
The purpose of creating an information-analytical system for processing large data is to release the doctor from routine work, which will monitor the patient's
condition as a whole and more effectively monitor the course of the disease. The main postulate of Big Data-projects is not to replace a doctor in matters of diagnosis, but to provide modern tools for effective work.
The strategic directions of application of Big Data in the field of health care are: creation of registers of medical data between which information exchange is possible; use of accumulated information to predict possible epidemic "outbreaks" of diseases; introduction of an electronic card for the patient, which will be available to every doctor who treats him.
Given the constant increase in the population and the proportional increase in the number of gadgets, processing large amounts of data is a necessity. Today there is too much information in the world, and every day there is more of it. For example, the audience of Google, YouTube, Facebook or Instagram is measured in billions of people. Even if we discard all the "bots" -let's try to imagine how many actions these people perform at the same time. In this case, Big Data (BD) is a log of user actions.
Such actions should not only be properly assembled, but also structured and analyzed. The main difference between BD and a conventional database is that information is received constantly, it becomes more and more, but it is necessary to analyze this data taking into account all the previous clusters of information.
The Internet of Things also produces a lot of such data: fitness trackers, smart watches, video surveillance cameras collect and group into clusters, information about users on a gigantic scale. But another problem for analysts lies in the fact that information about each user has many criteria.
The most popular areas of big data processing today in medicine are ophthalmology, dermatology and radiology. These are the areas of modern medicine where there are more imaging influences on diagnostics. Catapult, for example, uses big data analytics to help identify the relationship between injury and exercise, monitor exercise intensity, and alert individuals if they are potentially injured. Leading National Basketball Association (NBA) and National Hockey League (NHL) teams use Catapult services [5].
The use of AI in radiology facilitates workflows that can lead to a better and more accurate understanding of the disease. Prediction of cancer progression based on imaging data may become part of the clinical routine in the next few years [6].
Surgery is one of the most promising vectors for the development of BD in medicine. Even the most experienced surgeons do not want to operate on a patient who might die in a few minutes. Therefore, the system analyzes all available transaction protocols and visualizes risk areas. This allows the doctor to assess the risks of the operation much better. At the same time, information on all patients is collected in a large cluster. It is sorted by criteria and subsequently looking for addictions, confirming or refuting previous diagnoses.
Analysis of all known medical histories and diagnostics will allow to introduce into the practice of physicians a system of support for medical decision-making. Doctors will have access to the experience of tens of thousands of colleagues around the world.
Data from electronic medical records have already allowed doctors to establish a link between seemingly fundamentally different diseases. Developed in 2013 by the Kaiser Permanente consortium, the risk assessment system can predict the development of dementia in patients with diabetes. Using the same model, the US military is trying to reduce the number of suicides among war veterans [7].
Researchers at the University of Cape Town (UCT), after analyzing the most common types of cancer, concluded that each of these cancers is characterized by a clear combination of genes. According to the study leader, the team would not have been able to make the discovery if it did not have access to the Big Data arrays [8].
Children's Hospital in Toronto, Canada has implemented the Project Artemis project - a medical information and analytical system of the hospital collects and analyzes data on infants in real time. The system can monitor 1260 indicators of each child's condition every minute, allows to predict an unstable condition, and to start prevention of diseases in children in time, which significantly improved the quality of medical care and showed a significant economic effect [9].
Massachusetts General Hospital (MGH) physicians use the QPID analytical system to monitor important patient information during treatment [10].
Another application of the QPID system in health care is surgical risk prediction. The QPID system automatically searches for treatment protocols and then displays the results with a calculated red, yellow or green risk indicator [11].
In the spring of 2015, Apple and IBM announced a joint project to use Big Data in healthcare. The two corporations operate on a single platform that allows iPhone and Apple Watch owners to send information collected during use to Watson Health, IBM's medical analytics service [12].
About 35% of healthcare organizations plan to use artificial intelligence in the coming years. In this regard, Brendan FitzGerald, Director of Research at HIMSS Analytics, points out that the prospects for the use of artificial intelligence in medicine are quite high [13].
Big Data and Artificial Intelligence are now widely used in drug manufacturing and pharmaceutical marketing. The main areas of application of BD in pharmacy are the creation of new drugs; collection of clinical data on patients; improving the quality of clinical trials; modeling of new drugs, which is considered the most promising.
Big data is also used to predict side effects for specific compounds and components before clinical trials begin. By using an analytical method that involves testing dozens of different drug characteristics, companies can save time, money and save patients' lives.
B. FitzGerald notes that the ability to embrace the experiences of real patients, representing a wider sample of society than the relatively narrow choices included in traditional trials, is becoming increasingly useful as medicine becomes more personalized [13].
However, it should be noted that a large number
of experts in the field of big data believe that many projects using BD, and not only in medicine, fail precisely because of the presence of large amounts of unimportant information, so-called noise, which can lead analytical systems to erroneous conclusions.
An example of this is the Google Flu Trends Program (GFT), which was proposed as a method of assessing influenza among the general population and was used in conjunction with traditional epidemiologi-cal surveillance systems. Several previous studies have found that GFT scores were often inflated in real time, and a false relationship was found between disease onset and external factors. One of the reasons for these errors was to change the Google search tool itself, which led to the collection of disparate data [14].
Conclusions and prospects for further research. One of the important factors of unsatisfactory management in the field of health care in Ukraine is the low efficiency of information management in the health care system, which is due to the lack and isolation of Big Data registers and outdated filling of statistical forms; insufficient and uncoordinated use of modern information and communication technologies. In particular, data management in health facilities and statistical systems is in some cases unmanaged and in others excessively managed, resulting in unbalanced management of medical data, which results in results that are incomparable to the efforts made. The information space of the health care system is fragmentary, there is no operational data on most health care parameters.
This problem can be solved only by using a systematic approach to reengineering in the management environment of the medical field and the appropriate architecture of interaction. Transformational changes in the health sector must be effectively coordinated and have a reliable source of funding, and this requires reform of the financial support system, internal and external interoperability of medical information systems and registers with the introduction of compulsory health insurance.
References
1. Ethics And Governance Of Artificial Intelligence For Health. Digital Health and Innovation. Department Division of the Chief Scientist. World Health Organization Avenue Appia. Geneva, Switzerland. 165p.
2. Ali Nasser Al-Tahitah. Which one is more important to organizations; Human Capital Development (HCD) or Human Resources Management (HRM)? [Online], available at: https://www.re-searchgate.net/post/Which-one-is-more-important-to-organizations-Human-Capital-Development-HCD-or-Human-Resources-Management-HRM (Accessed 12 June 2021).
3. Framework and standards for country health information systems. Health Metrics Network, World Health Organization. 2nd ed. Switzerland. Geneva. 2018, 71 p.
4. Some issues of the electronic health care system: Resolution of the Cabinet of Ministers of April 25,
2018 № 411. [Online], available at: https://za-konrada.gov.ua/laws/show/411-2018-%D0%BF#Text (Accessed 14 June 2021).
5. Catapult, The Global Performance Technology Leader In Elite Sports Acquires SBG Sports Software, Expanding Its Client Base And Product. [Online], available at: https://apnews.com/article/technology-software-fcbf414e8addee2329890afbdf75828e (Accessed 12 June 2021).
6. Artificial Intelligence in Radiology. Empowering clinical decisions with AI. [Online], available at: https://www.siemens-healthineers.com/medical-imag-ing/digital-transformation-of-radiology/ai-in-radiology (Accessed 14 June 2021).
7. Behnam Kiani Kalejahi. Big Data Security Issues and Challenges in Healthcare. [Online], available at: https://arxiv.org/ftp/arxiv/pa-pers/1912/1912.03848.pdf (Accessed 04 June 2021).
8. The University of Cape Town's tradition of research excellence has led to great inventions and discoveries. [Online], available at: https://www.uct.ac.za/main/research/innovation-at-uct (Accessed 04 July 2021).
9. Carolyn Mcgregor. The Artemis Project: Pushing New Frontiers In Healthcare Analytics.
[Online], available at: https://www.can-health.com/2016/11/02/the-artemis-project-pushing-new-frontiers-in-healthcare-analytics/ (Accessed 04 July 2021).
10. Mass General uses predictive analytics for surgical decision-making. [Online], available at: https://www.healthcaredive.com/topic/health-IT/ (Accessed 08 July 2021).
11. Faiz Tuma. Operative Risk. [Online], available at: https://pubmed.ncbi.nlm.nih.gov/30335273/ (Accessed 10 July 2021).
12. Apple & IBM Team Up For New Big Data Health Platform. [Online], available at: https://www.linkedin.com/pulse/apple-ibm-team-up-new-big-data-health-platform-bernard-marr?trk=mp-reader-card (Accessed 10 July 2021).
13. Brendan FitzGerald. Research Director HIMSS Analytics. [Online], available at: https://www.himssanalytics.org/brendan-fitzgerald. (Accessed 12 July 2021).
14. Sasikiran Kandula, Jeffrey. Shaman Reappraising the utility of Google Flu Trends. [Online], available at: https://journals.plos.org/ploscompbiol/ar-ticle?id=10.1371/journal.pcbi. 1007258. (Accessed 12 July 2021).
