In recent years, medical care has moved away from the abundant experience and inspiration of veteran doctors, and the trend is toward making comprehensive judgments based on records of vast amounts of scientific diagnostic information, and then selecting the most appropriate and evidence-based treatment methods. I’m here. Medical big data is indispensable to realize this, and it is expected to help control medical expenses.
How to use medical big data
[Disease prediction/early detection]
In a 2018 report compiled by the Japan Medical Association called “Utilization of Japan’s Medical Big Data,” one example is that real-time medical big data is effective in disease prevention.
When a person undergoes a medical examination or treatment at a medical institution, a medical document is created that summarizes various data such as interview information, test results including images, and prescription drug names. In general, the more data available from patients with the same symptoms, the easier it is to make objective judgments by identifying the disease and determining the degree and progress of the disease. For example, by analyzing CT and X-ray image data with artificial intelligence, it is possible to search for similar cases from various past patient data accumulated in the system, enabling early detection and early treatment of diseases that are difficult for the individual to detect. will be By doing so, it can be expected to have the effect of protecting the patient’s own healthy life and increasing the happiness of living.
[New drug development]
There are three problems in new drug development. It takes a long time before a drug is released into the world and becomes a business base, it takes a huge amount of development costs, and the probability of successful development is low. The use of medical big data is attracting attention as a solution to these problems in new drug development. By narrowing down research targets that may become new drugs, and improving the quality of research data by analyzing it with AI, the success rate increases, and the new drug development process can be carried out efficiently and smoothly. As a result, new drug development speeds up and development costs fall accordingly.
[Recording medical records such as interview information and test results]
Even for the same disease, the treatment varies depending on the progress of the disease, physical strength, age, etc. of the patient. If medical records, which are medical records, are analog, the information is isolated without being shared anywhere. Electronic medical charts, which have rapidly spread in recent years, share medical information within each medical institution or within a system in which multiple medical institutions participate. , the possibility of providing optimal medical services to each patient increases.
In this way, the accumulation and appropriate classification of medical information related to symptoms, treatments, and results, as well as the construction of an interlocking system, are important themes. For example, a seemingly unrelated disease can affect other parts of the body. Medical studies have already shown that people with diabetes are more likely to develop periodontal disease, and that periodontal disease increases the risk of making diabetes even more severe. However, internal medicine and dentistry are different clinical departments, and currently it is difficult to share information. If the symptoms and treatment information of different diseases can be linked, it will be possible to provide comprehensive treatment, which is advantageous for both the treatment side and the patient side.
Current Challenges in Medical Big Data
Although many benefits are expected from utilizing medical big data, there are also challenges. It has long been pointed out that when the amount of data becomes enormous, it becomes difficult to manage, and it is difficult to retrieve the necessary information. However, with the evolution of data management systems and the introduction of AI, these problems are gradually heading toward solutions. In 2011, the digitization of receipts was made mandatory, and the national database managed by the Ministry of Health, Labor and Welfare is now used. Rather, the problem has shifted to how to protect personal information included in medical information. Medical charts created by doctors, which describe medical conditions, medications, surgery history, etc., are often classified and managed by patient name, and are important personal information that must never be leaked. In order to use medical data without problems, it is essential to anonymize personal information, eliminate the risk of information leakage or falsification due to unauthorized access, and be able to accurately pick up useful information. In order to collect and anonymize patient data from each medical institution, in December 2019, the Cabinet Office certified a business office whose representative director is a professor emeritus of Kyoto University. As a result, from the spring of 2020, various research institutes and pharmaceutical companies will be able to utilize medical big data. The use of medical big data is extremely useful from a medical point of view, but the first important point is proper information management.