Personalized medicine, regenerative medicine, genome. Extension of healthy life expectancy by medical technology, biopharmacy, etc. Advancement of prevention, advancement of AI, high performance of robots in the technology field
Utilization of AI and robots in dementia care and long-term care
Technology development for a society that does not cause dementia that can be prevented
About efforts in the field of cancer genomic medicine and artificial intelligence care / dementia
Nursing robots provide nursing care services. Smart laboratories. Health management utilizing AI. AI assists doctors ... Medical care and nursing care in Japan, which is facing the world's first rapidly aging society.
Utilization of AI. Dementia care. Self-support. Prevention of aggravation. Status of need for long-term care, etc. Maintenance / improvement. Health / medical care / long-term care. ・ Data such as rehabilitation / nutrition ... Brain science. Actuators that drive robots Nursing robots provide nursing care services. Smart workplaces. Health management utilizing AI. AI assists doctors ... Japan is welcoming the world's first rapidly aging society In the reform of medical care and long-term care
AI-equipped robots guide conversations in elderly groups to prevent dementia-Healthcare AI Development Acceleration Consortium
Supporting medical care and long-term care with AI x robot arm

"It was developed by combining AI (artificial intelligence), biomechanics, and ergonomics. It is a robot that has visual and tactile sensations and can imitate the delicate movements of human fingers.
Robot arms are generally used on production lines. If it is applied to long-term care or rehabilitation, it must break through the hurdle of having delicate movements like human fingers. The ultra-regional research team has developed a robot that combines AI (artificial intelligence), biomechanics, and ergonomics to have visual and tactile sensations and to mimic the delicate movements of human fingers. It can hold the ball and remove the tissue like a human. Designed to imitate the joints of human fingers, it bends the fingers by transmitting air pressure. In addition, the repulsive force received by the fingers by the arithmetic chip using the patented technology of the United States is measured by the sensor, and the survey is performed at an accurate angle. Generally, when a robot grabs a ball or grabs a bottle of beverage, it mechanically opens and closes its five fingers to grab or grab an object. However, it has a human-like smoothness. This is because human movements were digitized and the robots learned them little by little. If both hands can be moved like humans, robots will be able to hold babies, rub the backs of sick and old people, and raise their bodies in the future.
In addition to the delicately movable fingers, the team made a seven-axis robot with zero blind spots. This has a wider range of movement than a general 6-axis robot. You can bend your arms to an angle that humans cannot reach. In addition, because it has a freely movable neck and a 3D stereoscopic visual system, it is possible to scan surrounding objects in a wider range. In other words, it has a sense of "space." Moreover, it is possible to identify things by AI.
The part corresponding to the "brain" of the robot was researched and developed mainly by Professor Tatsu Kinnaka, Associate Professor Huang Ji, Associate Professor Zhu Hongkuni, Professor Seifumi Wu of Electrical Engineering, and Professor Yasushi Liu. Various breakthroughs in research and development that have won the "FutureTech Breakthrough Award" and the "National Innovation Award" have been applied so far.
What is the possibility of creating a combination of technology and "tactile sensation," which has many unexplained parts of the five human senses?
Listen with your skin. Haptics, a tactile information transmission technology, is attracting a lot of attention. In addition to the tactile reproduction already implemented in game controllers, various applied research such as remote surgery machines and alternatives to Braille is in progress. Facebook, which has launched an innovative approach in this field, is exploring the future of SNS that conveys symbolized emotions such as "likes" by stimulating the skin.
I have been studying tactile sensation at university for nearly 20 years. However, the general reaction is that we do not know what tactile research does. In fact, in recent years, a technology called "Haptic" (meaning "tactile"), which gives tactile feedback through vibration and movement, has been attracting attention, and Facebook is showing its presence in this field.
Research on human senses can be broadly divided into three fields.
One is the physiological field of investigating how people perceive the outside world using their senses.
The second is the field of sensors that artificially detect information with a mechanism similar to humans.
The third is the field of artificially presenting such a feeling.
Taking "vision", which is the most studied of the five senses, as an example, the physiological cells on the retina that are sensitive to the three primary colors of RGB (red, green, and blue) respond to each component of light. The first field is research on various parts and research that clarifies people's perceptions when they see various patterns of stimuli.
And the second is to make a camera by artificially realizing such basic performance.
The third field corresponds to making displays and projectors that present the images recorded in this way.
Tactile research is basically the same as these. "Elucidate the sensor of human skin sensation" "Realize sensing that gives the surface of the robot a tactile sensation similar to human" "Reproduce the sensation artificially by presenting an appropriate vibration / pressure pattern" Research is being conducted all over the world on themes such as.
Tactile research that is also used in Nintendo Switch
Then, in what fields can such tactile technology be used?
For example, the vibration of smartphones can be said to be the result of widespread tactile research. In recent iPhones, instead of physically pressing the button, by reproducing an appropriate vibration pattern, you can reproduce the feeling of clicking.
In recent years, game companies have been particularly focusing on tactile research, making use of it in controller functions. In Nintendo Switch, it is called "HD vibration" and appeals that it has a high definition (HD: High Definition) tactile sensation. With software called "1-2-Switch," there are games that you can enjoy just by vibrating, making use of the tactile sensation of the controller. DualSense, the PlayStation 5 controller, is talking about new features such as "haptic feedback" and "adaptive triggers." The tactile sensation is utilized in the form of generating a force that pushes the button back, improving the immersive feeling in the game.
An example of the research stage is an application such as a remote surgery machine. In endoscopic surgery, operations such as removing a tumor by controlling forceps from a distance, relying on images, are performed. At that time, by tactilely feeding back the force applied to the forceps, it is expected to have effects such as intuitively understanding whether other organs are damaged and improving the operability of the forceps. NS.
These are examples of reproducing the tactile sensation felt by humans, but there is also an application of transmitting symbolic information to humans via a channel called tactile sensation.
Braille is the easiest way to convey symbolic information by touch. The blind person can understand the character information by touching the uneven pattern through the tactile sense. However, there are not so many visually impaired people who can read Braille, and a 2006 survey by the Ministry of Health, Labor and Welfare reported that the percentage of visually impaired people who answered that they could read Braille was 12.7%. Braille is not always indispensable because some visually impaired people have low vision such as amblyopia and narrowing of the visual field, and more recently, machine-based text-to-speech technology has become widespread. However, the penetration rate of Braille as a visual alternative information acquisition channel is by no means high.
About this, some research has been done on whether it is possible to convey textual information via tactile sensation without training. A previous study in our laboratory found that if you hold a pen and the pen moves to write letters automatically, the person holding the pen can recognize the text information with considerable accuracy. .. It has been confirmed that the 26 letters of the alphabet can be understood with an accuracy of 90% or more at a pace of one letter per second.
"Typing with the brain and listening with the skin"
Facebook is actively researching the transmission of symbolic information via such tactile sensations. At the world Haptics, the largest international conference in the field of tactile sensation held once every two years, the company ranked first in the 2019 paper acceptance ranking by institution. The number is twice as large as that of the University of Tokyo, which ranks second. At the 21-year conference held the other day, although the number of presentations was small, probably due to the influence of the corona disaster, there is no doubt that he is focusing on tactile research.
Facebook first received a lot of attention in tactile research with the two approaches of "typing with the brain" and "listening with the skin" advocated in 2017. If we think of human beings as a "system that inputs and outputs some information," most of the information output is via voice, fingers, or a keyboard. On the other hand, most of the information input is audiovisual. It is either input through the eyeball by a two-dimensional color and light / dark pattern, or input as sound through the vibration of the eardrum.
Facebook's idea is to input and output this information on a different channel than before.
Regarding the "output" of information, it seems that they are thinking of reading the brain signal directly and outputting it as information without making a voice or moving their hands. This is the aforementioned "typing in the brain", but it is also a topic that has been widely studied in the field of Brain Machine Interface (BMI). For example, for patients with diseases such as ALS (Amyotrophic Lateral Sclerosis) that make it impossible to move the muscles of the whole body, there is an application that sends out one's intention to the outside just by thinking in the head. It has been considered. Recent advances in artificial intelligence (AI) and machine learning technology have made it possible to appropriately analyze measured EEG patterns. Around the same time as Facebook's proposal, Elon Musk, the founder of Tesla, a major US electric car maker, is proposing "Neuralink," a capacity expansion by implanting electrodes in the brain.
On the other hand, "input" strongly reflects our lifestyle as to which of the two information channels, visual and auditory, is used. For example, in the United States of the automobile society, information transmission using voice channels such as podcasts is popular. When driving, only hearing is available, and we are aiming for that. On the other hand, in urban areas of Japan, many people use trains when commuting to work or school. Since it is safe to occupy the visual sense on the train, it is more efficient to obtain information visually such as characters and images.
Facebook's aim is to use tactile sensation as a new channel here. Tactile sensation is a sensation that is widely distributed throughout the body. Since the four senses of sight, hearing, smell, and taste have localized sensory organs such as eyes, ears, nose, and mouth, it is difficult to present further information if they are occupied by some stimulus. .. In the case of tactile sensation, even if the handle is held by hand, there are still many unused areas such as arms, shoulders, abdomen and thighs. By giving an appropriate stimulus here, for example, if you can convey information as if you were listening to voice with your ears, you may be able to check the title of the incoming mail even while driving a car and listening to music. be.
Do you convey your feelings just by touching?
When I read the article "Listen on the skin" in 2017, I was most interested in how much information could be transmitted through the skin. I thought that the sense of touch was not smart enough to convey precise textual information. Opening of the anime "Lupin III" There is a production in which subtitles are displayed character by character like a flash, but many people may find it difficult to read. Similarly, when the tactile sensation was displayed symbolically one by one, I thought it would be difficult to connect them in the time direction and understand them as sentences.
The results of Facebook announced at World Haptics in 19 were exactly the answers to this question based on two years of research. In a word, the content was that it did not assume the transmission of precise text information.
So what were you trying to hear on your skin? It is "emotion". In one study, one subject was instructed to keep his arm out and another subject to touch her forearm to convey a particular emotion (such as joy or sadness). At that time, I was evaluating how much the person who was touched could understand the emotions. In other words, it is a study of whether feelings can be conveyed to the other party only through the sense of touch. An analysis of how they were touched during such movements and a study measuring how the nerves of the person touched at that time were ignited were also introduced.
In order to convey emotions just by touching them, there were many studies on hardware design, mainly targeting the forearm and what kind of device should be used to convey the necessary and sufficient emotions.
In other words, in order to realize "listening with the skin", 1) present stimulation to parts other than fingers such as wrists and forearms, and 2) convey emotions such as joy and sadness depending on the stimulation method. I am aiming for.
I think this makes a lot of sense for Facebook's strategy. On SNS, users communicate with symbolized emotions such as "like" and "sad." It would be nice if we could intuitively feel these several kinds of emotions.
If you think about it, even with LINE etc., you often end up with stamps without writing comments directly. If you can somehow convey the emotional nuances that accompany the stamp, you don't need a sentence there. A series of research by Facebook made me feel that the use of tactile sensation as a new channel for that purpose will come out in the future.
Can we fill our senses with excess information?
In writing this article, the editor asked me, "Is it good for people to fill up their senses more and more?" To be honest, this was a perspective I had never thought of, and I was surprised.
Information-related researchers who handle human senses basically have the issue of how to utilize the vacant senses. With the spread of smartphones in recent years, audiovisual is being occupied for a considerable amount of time. For example, we take our smartphone out of our pocket, even when we're up a few floors in the elevator. The tendency to fill audiovisual with information is certainly increasing year by year.
Facebook's strategy is to use tactile sensation so as not to compete with audiovisual sensation that fills up quickly. It is reasonable from the viewpoint of information utilization, which is to properly use the vacant sense, but from the viewpoint that "the sense is being filled more and more", it may be necessary to stop for a while.
In recent years, the increasing number of people who are attracted to meditation called mindfulness may be understood from the perspective of opening up a buried sense. The content that fills the senses, the new tactile technology that utilizes the senses of the gaps, and the ingenuity to release the buried senses are similar to eating supplements while eating a high-calorie meal and working at the gym. Can you say that? We are in an era where we can't stop even if we know that we shouldn't get so much information from the beginning.
Finally, I mentioned that "tactile sensation is a sensation that is rarely used in the mainstream at present", but there is one case where the author feels that this is the ultimate communication that can only be achieved by tactile sensation. .. It is communication with the pregnant foetation. Through the skin of the pregnant woman's abdomen, the vibration of the foot kicked by the foetation feels life, and if there is a reaction to knocking, it will be happy, which is an impression that can only be experienced by touch. There is no abundant and high-density information like audiovisual. Ambiguous, sometimes one-way vibration-only communication. It would be nice to be able to artificially design the luxury of filling time with such sparse communication, but it cannot beat the power of life that is born.


Medical expenses and the elderly population
The World Economic Forum cites aging as one of the five main drivers of global change, along with widening income-wealth inequality and climate change.
 
Global cost
The world population over the age of 65 will triple over the next 30 years to 1.5 billion, and the costs associated with aging are expected to rise accordingly. In fact, some estimate that by 2050, the economic impact of chronic diseases such as heart disease, Alzheimer's disease and diabetes will exceed US $ 47 trillion.
Increasing medical costs in the US and UK
Under the US healthcare system, people over the age of 65 spend more than five times as much on health care as under the age of 25. By 2030, the United States is expected to have three working populations per person over the age of 65.
Japan
Currently, medical expenses account for 8% of GDP in Japan, but research shows that it is the world's oldest society due to decades of economic stagnation, excessive use of expensive branded medicines, and so on. It is estimated that medical costs will increase to 11% of GDP by 2025.
While the elderly population accounts for 28.1% of Japan's total population, 4.6 million people, or one in five elderly people over the age of 65, have Alzheimer's disease, which is a serious epidemic. Facing This number is expected to increase as the elderly population grows.
Japan has taken a major step as a world leader in the practical application of drone use in the regenerative medicine, robotics and AI, self-driving cars, and service industries. All of these are designed to play an important role in achieving superior mobility, connectivity and longevity for an aging society. In particular, Japan is promoting the growth of support robots. Health care is an important pillar of Society 5.0. Japan's health care is supported by the universal health insurance system, and the wealth of raw data from horizontal connections between academia, industry and government has laid the foundation for Japan's economic model for decades. ..
To solve Japan's aging problem, the Japanese government is supporting the introduction of robots into nursing care facilities. However, many strongly recommend a careful balance between long-term care benefits and ethical issues.
Aging
• Genome instability: Both internal and external factors that cause gene damage accumulate in the body throughout life. This has been shown to accelerate aging.
• Telomere shortening: Telomeres are "lids" that protect the ends of chromosomes and shorten each time a cell divides. Over time, telomeres shorten until cells can no longer divide, which can lead to illness.
• Epigenetic mutations: Individual life experiences and environmental factors cause mutations in gene expression rather than DNA, which affects aging.
• Loss of proteostasis: Cellular proteins are misfolded with age, resulting in loss of homeostatic function. Accumulation of such damaged proteins causes aging and aging-related illnesses.
• Nutrient sensation dysregulation: There are pathways that regulate metabolism, the proteins of which are affected by nutritional levels and are also involved in accelerating aging.
• Mitochondrial dysfunction: As we age, mitochondria (the energy factories that regulate metabolism in the body) begin to dysfunction.
• Cellular senescence: “Aged” cells cannot be eliminated as quickly as younger cells and can accumulate in the body, which can adversely affect health.
• Stem cell depletion: The activity of four types of stem cells that help regenerate new tissue cells declines with age.
• Changes in cell-cell communication: Cell-cell communication is impeded with age, causing inflammation and tissue damage.
Aging is a natural feature of biological organisms, and most advances in AI and deep learning in the field of aging research are in the area of ​​expertise in biomarker development. Age prediction systems, or "aging clocks," have been developed for a variety of data types, from basic clinical blood tests, photographs, videos, audio, retinal scans, and even medical imaging to microbiome data.
Digital medical market
Healthcare is a $ 3 trillion US industry, and artificial intelligence will create new and innovative megacorporations that will dominate in the 2030s and 2040s.
In 2018, venture capital invested $ 10 billion in nearly 500 startups in the United States, according to CBinsights. Leading technology companies, venture capital firms, and governments are also enthusiastically investigating technological advances.
Investment stage of digital health start-up companies
There are 18,000 start-ups in the digital healthcare market, worth $ 145 billion. Investors invested $ 6.8 billion in digital health in 2018. AI regulations are not yet in place, but digital health will be more or less fully implemented and will be easier to implement. The medical AI market is expected to achieve a combined annual growth rate of 40% by 2021. This is primarily because AI has the potential to improve medical outcomes by 30-40% while cutting treatment costs in half.
The field of digital health is expanding rapidly. From virtual diagnosis to the development of more effective medicines using AI, the possibilities are expanding dramatically.
Trends-AI impacting medical and health data in unexpected ways
From early detection to improved medical diagnostics, AI continues to transform the world. AI and machine learning are transforming health care in many ways, including how consumers use services, how healthcare providers provide health care, and what health outcomes can be achieved. AI has destructive power, and this is just the beginning.
Healthcare could eventually become the area of ​​greatest change with AI. Machine learning will have a major impact on areas such as medical imaging with neural networks, electronic medical data mining with NLP, deep reinforcement learning and robotic surgery using 5G. In addition, by applying machine learning technology to wearable devices, sensor data can be processed and used for preventive medicine.
2020 is expected to be the year when AI technology will be applied to medicine and pharmaceuticals. Humans will have the opportunity to modify, improve and continually improve their abilities and spirits.
New mobile technologies that monitor key biological indicators such as heart rate and blood sugar bring fundamental and positive changes to consumer behavior.
Investors have invested heavily in wearable devices that support weight loss, but in the end, no long-term behavioral changes have been seen, resulting in disappointment.
AI can have many implications for approaches to health, from prevention to self-monitoring, diagnosis and treatment. This also includes how to deal with the social, family, economic and emotional aspects of health.
In the future, data collected from social media sites such as Twitter and Facebook, wearable devices, environmental sensors, etc. will be combined with genetic and proteome data to more clearly define both the risk and onset of a particular disease. Will be important. Only by using this knowledge can companies use medical institutions. And consumers will be able to develop low-cost interventions that can be used to interrupt this process. Combining data collected through multiple diagnoses, traditional laboratory tests, and wearable devices provides a more comprehensive picture of disease risk, facilitating long-term behavioral changes for healthcare organizations and health coaches. You will be able to prioritize easy and low cost interventions.
AI and aging
The cost of treating aging-related illnesses can disrupt healthcare systems around the world, filling corporate and government agendas with this. AI can put an end to aging and generate agile initiatives that can adapt to the constant changes in society. We have made great strides in reducing infant mortality and reducing the number of deaths from infectious diseases. However, there are still older people with serious health problems such as cardiovascular disease, type 2 diabetes, and Alzheimer's disease, and millions of older people live lonely.
Human life has been extended, but quality of life has not improved. Advances in AI are having a major impact on the field of aging research. AI has the unique ability to identify relevant patterns in complex nonlinear data without a deductive or mechanical understanding of biological processes. AI reveals the complex, holistic and integrated relationships that are taking place in the body.
AI has great potential to transform how we age, monitor ourselves, and live our lives together.
The use cases that have been applied show that AI is beginning to produce promising results in various fields of research on aging and longevity. AI is gradually changing from an over-promoted technology with few proof-of-concepts to a large-scale adoption and recognized trend in geriatric medicine.
AI is already making rapid progress to eliminate process inefficiencies and expensive manual treatment problems and prevent human error, ensuring that it redefines the original purpose of patient care. doing. Unleashing the power of AI requires close collaboration with healthcare IT professionals and industry end users. We need to focus on building more complex social processes in order to improve human abilities and social connections. These mechanisms include new technologies such as self-driving cars, invisible IoT sensors, mobile health apps and digital health apps, as well as related services built to support economic growth.
We have already succeeded in developing AI algorithms in various fields of pharmaceuticals, and they are applied in a wide range of applications.
Regulations that encourage the integration of services into a common platform will drive innovation for healthy aging. Access and regulation need to take into account the need for collaboration between healthcare organizations and AI-based companies to establish an effective pipeline for data acquisition.
To take advantage of the useful information contained in this large amount of data, it is also necessary to develop deep learning techniques and introduce new computational approaches such as machine learning. AI's ability to make accurate predictions is highly dependent on data availability.
AI technology and use cases
Medical diagnosis
"Approximately 10% of patient deaths are due to misdiagnosis, which accounts for 6-17% of hospital problems." Source-National Academies
More than 12 million serious misdiagnosis cases occur each year, and there are many areas where the quality of medical care is not very good.
AI solutions can speed up diagnostics and improve accuracy. This allows for better diagnosis in multiple areas than most doctors do today. However, the integration of AI into medical practice is still in its infancy, and providing our most personal data to companies that are already collecting personal data for their own benefit poses many risks. I am involved.
Doctors' abilities are not the only factors that cause such mistakes. There are many other factors, including:
• Inefficient coordination and integration of health information systems
• Communication gap between clinician and patient
• Traditional medical services system with insufficient diagnostic support
Due to the limited data available on patient outcomes, the application of AI to medical diagnostics is currently in its infancy. By 2022, AI may be more likely to influence how healthcare institutions and healthcare systems approach diagnostics. AI will play a role in transforming an individual's ability to see changes in their health in real time.
Hospital management
Predict patient outcomes and use them for hospital management. By using AI to rearrange electronic medical records, you can reduce the huge waste of time required for medical management. This means that companies such as Amazon and Google are using smarter AI to repackage their medical data. AI has the potential to save a lot of time by streamlining documents and processes such as electronic medical records.
Drug development
According to a basic overview of HIMSS Analytics 2017, less than 5% of healthcare organizations are currently using or investing in AI technology.
Over the last five years, the introduction of AI has led scientists to rethink new drug development and disease control methods in the pharmaceutical industry.
The IT currently used by major pharmaceutical companies is traditional, based on legacy systems, non-interoperable, and untagged with data. With the introduction of AI-based systems, pharmaceutical companies can not only solve these challenges, but also reduce costs and develop effective new therapies.
AI will play an important role in the pharmaceutical industry in developing new drugs, supporting medication compliance, and in-depth analysis of clinical trials. Machine learning algorithms are also expected to be used to investigate the biological and chemical interactions of pharmaceuticals.
Medical Imaging and Radiology
The application of AI to medical imaging can enhance a wide range of essential processes such as medical image reconstruction, noise reduction, quality assurance, segmentation and triage.
Deep learning can primarily identify specific features of an image, improve image quality, and find outliers and anomalies. Many imaging laboratories are rapidly transitioning to advanced technology to bring efficiency and expertise to the ideal level.
Many of the future AI-based applications are said to have potential in the areas of radiogenomics, computer-aided detection, and classification. The current healthcare system will be rebuilt by AI, and as a result, current clinical diagnostic imaging will be significantly affected. However, you should focus on building a new pre-trained model architecture customized for medical imaging data, along with a seamless interoperable data exchange method.
In addition, the rapid advances in AI in radiology can be a decisive breakthrough in efforts to transform overall patient care. AI enhances an integrated cloud-based platform that enables radiologists to automatically review cases in real time.
Pathology-early detection of illness
With the rapid spread of digital pathology using AI, traditional pathological diagnosis is nearing its end. With increasing workload and a growing need for accuracy, AI will show its full potential in the coming years. State-of-the-art technology has the power to realize pathological diagnosis that cannot be achieved with a microscope or human vision.
The application of AI to pathology simplifies image analysis, identification of rare objects, morphological segmentation, and digital imaging of entire slides. The rapid introduction of artificial intelligence and digital pathology in recent clinical practice has opened up new possibilities for value-based care delivery.
IoMT --Internet of Medical Things
Combining AI and IoT can create great value. This advanced technology, called the Internet of Medical Things (IoMT), allows a variety of medical devices and sensors to connect to the Internet to collect vast amounts of critical patient data. These data collected can be analyzed and used to understand the patient's condition, make quick and accurate medical diagnoses, and understand resource usage patterns in medical institutions.
Many medical institutions are interested in the benefits of IoMT. IoMT provides great relief for patients with chronic illnesses and medical institutions. With IoMT, patients can be monitored in real time while at home comfortably.
Robotics
While 68% of those surveyed said that robots are beneficial because they can help humans, they are reluctant to get robot assistance or communicate with them in hospitals and long-term care facilities. Only 26% said they didn't have one.
Source: European Commission Survey-ec.europa.eu/digital-single-market/en/news/attitudes-towards-impact-digitisation-and-automation-daily-life
With the increasing number of older people in need of care, researchers believe that robots will be one way to meet the overwhelming demand.
• Interactive AI Agents-Interactive / cognitive AI agents such as Ipsoft Amelia for senior management and aging.
• The Robot Concierge helps older people continue to live alone. These robots provide all kinds of support, from cooking to cleaning to socializing. Robots provide long-term care and, in some cases, social interaction, but there are also challenges faced by some researchers, such as how to apply ethics to this unique human challenge. ..
• Mobile Robotic Telepresence (MRT) systems have already proven to enable constructive social interaction with older patients. The MRT is basically a wheeled robot with a head-high display that can be remotely controlled using a simple smartphone app. MRT allows relatives and social workers to frequently "visit" older people, even if they live far away. Elderly patients do not need to operate the device and can interact with social workers and family members without doing anything.
• Humanoid robots have evolved enough to provide the care that is especially needed by the elderly. Not only can these robots pick up objects and move independently, but they can also use, for example, the movements of their arms and hands to perform natural, human-like interactions. More advanced humanoid robots are equipped with devices and sensors such as touch screens. Many older people find the touch screen difficult to use and prefer to give the robot interactive commands and read the response from the screen. On the other hand, the option to use a touch screen is essential for the elderly with deafness and visual impairment due to aging. Humanoid robots have focused on assessing the functionality of technology without considering the potential social and ethical implications. Due to its nature, humanoid robots are generally thought to help reduce loneliness.
• Surgical robots will be able to perform better than doctors in many routine surgeries in the near future. The introduction of medical technology is expected to accelerate further in the future.
Deep learning neural network
Deep learning makes inferences based on previously encountered data is. For example, a great idea of ​​how AI will affect almost every conceivable diagnosis, process, outcome, and physician, not just medical, with more medical-related data. Can be obtained.
The latest deep learning techniques used in the development of age prediction systems offer new possibilities for previously incompatible dynamic and static data types. A new method for building a causal model that uses AI-powered aging biomarkers to gain a complete picture of biological processes, extract the most important features, and identify biological targets and mechanisms. Will be possible. Recent developments in the Generative Adversarial Network (GAN) and reinforcement learning enable the generation of diverse synthetic molecules and patient data, the identification of new biological targets, and the generation of new molecular compounds with expected properties and gero protectors. became.
These new technologies combine with a real-world evidence pipeline that may help accelerate and improve integrated seamless end-to-end biomarker development, target identification, drug discovery, and drug R & D. I can.
 
Advanced use of deep learning to prevent aging
Precision medicine by discovering aging biomarkers
Precision medicine relies on robust quantitative biomarkers. Aging biomarkers are tools that can provide a quantitative basis for assessing the therapeutic effect of clinical interventions.
One of the major challenges facing human aging research today is the lack of biomarkers that can be targeted and measured to track the effects of therapeutic interventions in anti-aging.
Currently, many aging biomarkers have not only one function, but a limited set of physiological functions that are known to induce the development of certain aging-related diseases and dysfunctions when destroyed. Monitor Therefore, there is still a need to develop biomarkers that are characteristic of biological aging that are objectively quantifiable and easy to measure. Designing such biomarkers requires a time-consuming and tedious multi-step process that includes proof-of-concept, experimental verification, and analytical performance verification. AI technology provides an effective alternative to the development of aging biomarkers. Data-learning-based aging clocks have already been used to identify biomarkers. In the future, complex data learning architectures will be used to unravel complex non-linear relationships between data features and build even more accurate models.
Target identification
Another important aspect of developing effective anti-aging therapies using deep neural networks is the identification of targeted targets. Various computational approaches have been developed, including screening for differences in pathway activation patterns using pathway perturbation analysis. These techniques can be used to characterize pathways as transcriptome maps and identify pathways that cause major changes between young and old. The results of such analyzes provide information about the pathways involved in aging. This technique, which uses deep neural networks, relies on a screening library of known compounds identified using potential longevity properties. You can also analyze the features used by the aging clock to identify new targets. This approach uses a list of genes used to predict age based on skeletal muscle characteristics and further analyzes it to identify the genes most important for age prediction. Several techniques have been used to assess the importance of features in age prediction.
Small molecule drug discovery
Computational methods are being developed to design drug compounds and build large virtual chemical libraries that can be screened more efficiently in drug discovery. Drug discovery and drug development can be further optimized by using deep learning technology to characterize drug candidates according to their potential efficacy and safety. AI technology disrupts the standard drug discovery pipeline, enabling effective identification of new candidates for anti-aging treatments. For example, a generative model based on the GAN paradigm has great potential due to its ability to generate virtual molecules.
Gene therapy
Gene therapy is an experimental technique that uses genes to treat or prevent diseases such as hereditary diseases, various cancers, and certain viral infections. In fact, this technology is designed to supplement abnormal genes and produce beneficial proteins by introducing genetic material into cells. Although gene therapy is promising, it is currently being tested only for diseases for which there is no other cure. The FDA has approved gene therapy for the first time in the treatment of some leukemias, lymphomas, and the hereditary disease retinal dystrophy. Artificial intelligence is an important factor in improving the accuracy of the gene editing process. The development of gene editing opens up new opportunities in the search for personalized therapies by providing scientists with the potential to alter a patient's DNA. With the ability to learn, predict, and advise on vast amounts of data, AI technology offers a variety of options for predicting patient prognosis, from experimental clinical trials to tailor-made medicine. You can identify patterns that you can use to advise your doctor. However, there are still challenges as to whether genes can be edited accurately.
Leading Technology Company – FAGMA
Google, Alphabet, Microsoft, IBM, Amazon, and Tencent are all entering the medical and medical technology arena. All of these companies are taking big steps to connect their future with AI in the medical field. These leading tech companies are pursuing data and working on standards and toolkit designs.
To protect the world from a variety of illnesses that could affect the global economy and the rapid aging of the world, tech companies are trying to put a stop to medical costs.
As software is a differentiator in healthcare, leading tech companies are in a better position with their software expertise, direct consumer relationships, and ample funding.
A pharmaceutical company affiliated with a biotechnology AI startup for drug discovery
Traditional pharmaceutical companies are expanding their partnerships with AI SaaS start-ups in search of innovative solutions to long drug discovery cycles, as well as with a variety of biotechnology AI start-ups.
Pharmaceutical companies are investing millions of dollars in AI algorithms to discover new treatment candidates and transform the long-term drug discovery process. At the same time, we use digital health to reach consumers directly and find the necessary expertise in AI.
Regenerative medicine and AI
Regenerative medicine is an area of ​​greatest potential for entrepreneurs with the ability to generate and grow revenue. Human life is increasing and people want to live as high quality as possible for as long as possible.
The purpose of the field of regenerative medicine is to provide patients with more advanced treatment and shorten recovery time, for example, by using artificial stem cells. AI technology offers the opportunity to develop predictive models for personalized medicine using artificial stem cells, immune cells, and human regenerative tissues. For example, AI can be used to identify the developmental state of germ cells. Another application of AI technology has recently introduced an online catalog of 3D stem cell images generated using deep learning analytics and cell lines modified with the gene editing tool CRISPR.
To a greater extent, AI can provide realistic solutions to other types of challenges facing regenerative medicine technologies, as well as computational model-based deep neurals for more precise tissue engineering applications. With the development of networks, it is possible to predict the results of tissue engineering.
A start-up company focusing on personalized medicine and anti-aging research realized by AI
AI is used to develop effective personalized medicine based on the interpretation of large medical data sets collected by businesses and healthcare organizations.
Advanced biometrics
Researchers are using neural networks to investigate and measure risk factors that were previously difficult to quantify. AI's ability to discover patterns will continue to pave the way for previously unknown risk factor identification and new diagnostic methods. Analysis of retinal images and audio patterns using neural networks may help identify the risk of heart disease.
IoT-Smart Device Enterprise
AI as a medical device, software, service
In the last few months, many AI software products have been approved for clinical commercial use by the FDA. Due to the rapid progress of regulatory approval, a total of 149 products have been approved by more than 80 AI imaging and diagnostics companies that have expanded their investment since 2014.
Several AI-as-a-Service (AI as a Service) platforms are integrated into FDA-approved home surveillance devices to alert doctors in the event of anomalies.
Robotics
We will create a companion robot that is programmed with advanced artificial intelligence and can interact with humans independently.
Overview of other areas
Telemedicine also improves care by facilitating communication between patients and doctors. Telemedicine offers a better opportunity to treat and monitor chronic illnesses 24 hours a day. Biomedical telemetry will make great progress in the future. This digital technology has the potential to expand its capabilities in the area of ​​assisting physicians and relieving stressful situations. Bioengineering and biomedical equipment will introduce new solutions for people with disabilities or those suffering from serious illness or injury.
Prostheses are being developed at a very high rate. Lightweight and reliable new materials, 3D printing technology, and smart algorithms enable the development of advanced prosthetics.
Smart Homes-Healthcare in smart homes is expected to grow explosively in the future, and with the maturity of fast-growing technologies such as 3D printing, many companies have the advantage of dominating this area. I am in a good position.
AI-enhanced longevity medicine
Companies working on longevity drug development are more advanced than traditional biopharmacy companies. Longevity research and development uses many scientific and technological fields such as AI, digital medicine, and jeroscience, so companies working on longevity are considered next-generation companies. Longevity companies are using next-generation technologies to conduct clinical trials, as well as highly sophisticated techniques for diagnosis and prognosis.
Senolytics. One of the most popular areas of research is senolytics. It is a class of medicines that target senescent cells and destroy them by inducing cell death.
Senolytics is part of a new field of study, including jero protector drugs that are thought to be able to prevent or rejuvenate aging by targeting cell stimuli such as DNA damage. It will be.
The following are the major companies involved in clinical trials.
Metformin
Metformin is an inexpensive generic drug for the treatment of type 2 diabetes and has emerged as a drug with the potential to promote longevity.
Longevity researchers are looking at its ability to reduce oxidative stress and inflammation associated with aging.
Genome instability refers to mental stress, solar radiation, harmful chemicals, and biological factors that directly cause DNA replication errors or induce DNA damage through the activity of reactive oxygen species. Our genome is potentially damaged every day Facing The ability of the genome to maintain integrity and stability is due to a unique repair system that repairs this accumulated damage.
Epigenetic mutation
The process of aging can cause epigenome changes. This change can potentially result in altered gene expression that can potentially alter cell function and ultimately impair that function. As an example, epigenetic mutations in the immune system can inhibit activation, suppress immune cells, and make humans vulnerable to immune system disorders and inflammation.
Loss of protein homeostasis
Loss of protein homeostasis is a common process associated with aging and is an indicator of disease. This process is characterized by the accumulation of harmful or misfolded proteins and unwanted aggregates. Degradation of these proteins is inhibited with age. It has been found that aged cells have reduced autophagy and cell lysosomes are less efficient at removing cell waste products. Alzheimer's disease, Parkinson's disease, Huntington's disease, and prion's disease are examples of diseases caused by loss of protein homeostasis.
Insufficiency of nutritional perception
Metabolic activity is known to stress human cells. Calorie restriction is a preventative therapy to delay metabolic dysfunction. Some metabolic by-products, such as mitochondrial dysfunction and oxidative stress, can cause diabetes by losing control of nutrient sensing. There are several important metabolic signaling pathways that can regulate the rate of aging. Proteins may be the subject of anti-aging research. Intermittent calorie restriction remains the most widely recognized prophylactic therapy to delay potential metabolic dysfunction.
Mitochondrial dysfunction
Not many drugs are under development for mitochondrial dysfunction. Mitochondria are involved in the metabolism of urea, cholesterol and neurotransmitters, detoxification and free radical production, as well as digestion of fats, carbohydrates and proteins. These dysfunctions cause many systemic disorders, and many common aging lesions have been shown to result from mitochondrial dysfunction. Mitochondrial dysfunction is associated with type 2 diabetes, atherosclerotic heart disease, cancer, Alzheimer's disease, Parkinson's disease, and stroke. Mitochondrial dysfunction is considered to be one of the main reasons for human aging.
Cellular senescence
As we age, cells that die and refuse to replace accumulate. These cells inhibit cell function, reduce tissue repair, accumulate mutations that can cause cancer, and cause an increase in chronic inflammation. This phenomenon, in which the cell division cycle is irreversibly stopped, is called cellular senescence. Aging is a universal process that affects almost every cell. With age, the immune response declines and some senescent cells escape immune clearance. By the time humans age, a large amount of senescent cells have accumulated. The long-term presence of aging cells in tissues can accelerate aging-related illnesses and even accelerate aging itself.
Stem cell depletion
Stem cells differentiate into many other cell types and can function as regenerative components in tissues. Stem cells replenish tissue with functional differentiated cells to maintain normal cell and tissue function. The number of stem cells continues to decline throughout life, so younger people are more likely to be replenished. Stem cells tend to lose their ability to differentiate with age because they accumulate genetic variation and damage with age. Stem cell depletion represents a failure of tissue regeneration and is characteristic of aging.
Changes in cell-cell communication
As humans age, the electrical and chemical signaling systems between cells are disrupted, resulting in signal distortion. The destroyed message results in an imbalance in cell-cell communication. As cells age, more inflammatory signals are produced due to the dysfunction of the immune system and the accumulation of tissue damage. The result is chronic inflammation and chronic diseases such as atherosclerosis, type 2 diabetes, obesity and chronic peptic ulcer.
A New Future-Evolving Humans for an Evolving World-Intelligent Society
We will see increased use of augmented reality, increased cross-border communities, and significant growth in digital enterprises. All of these trends represent the evolution of human digitalization and new aspects of the world economy.
When it comes to general-purpose AI, there are several efforts being made, such as the development of general-purpose artificial intelligence (AGI) and the transfer of human memory and abilities to computers. Concepts that show the feasibility of these approaches. No proof has been done.
The birth of hybrid humans is just around the corner. Perhaps the most amazing ability of AI in medicine is the overlapping of technology and human ecology.
Advances in AI are approaching the day when chips embedded in the body can control internal autonomous systems. These chips are activated by reactions in the brain, where virtual reality enhances vision and enhanced storage capabilities enhance memory.
By embedding AI chips in the human brain, we can also have a clear path to unraveling the double helix. It allows you to write and edit human DNA, remove congenital defects, and correct illnesses and illnesses as soon as they occur.
We bring the latest AI capabilities to generalize, learn strategies, and generate new models, targets, and data from learned features, as well as new applications and solutions in all areas of preventative, regenerative, and restorative medicine. We need a multidisciplinary approach that combines accurate methods of feature extraction and causality analysis.
Simply put, the latest advances in big data, software, chemistry, biotechnology, and machine learning can enable additional human capabilities that are uploaded directly to the human brain. In the future, synchronizing human ecology with the latest technology will open up entirely new and unknown territories. The possibilities are endless.
AI research on aging shows the need for consortiums and international cooperation focused on the economic aspects of data, generation, exchange, model exchange and validation, and analysis, clinical trials, and educational programs.
It has been said that the field of long-term care is a field in which the use of ICT is delayed compared to other industries, but the movement to embark on reforms is gradually becoming active with various issues accompanying the aging of the population. The spread of new normals accompanying the spread of the new coronavirus infection has triggered a review of the way long-term care should be, and the national government has begun to build a long-term care database with the aim of realizing "scientific long-term care" based on data. Be done. In this paper, we will introduce the latest trends toward ICT utilization in the long-term care industry, along with background factors and a sense of challenges.
ICT utilization area in the long-term care industry
There are structural issues in the long-term care industry behind the urgent use of ICT in the long-term care field. One is the growing shortage of human resources. The population requiring long-term care is increasing due to the aging of the population, but the supply of long-term care workers who support it has not caught up, and the output gap is widening. The work burden on field staff is heavy, and there is an urgent need to streamline operations with limited care personnel. Another issue is that the business environment of long-term care providers is deteriorating due to the tight financial situation of long-term care. In order to improve and strengthen management, it is required not only to reduce costs but also to develop and provide new added value that leads to profits.
Currently, the main ICT utilization cases seen in the long-term care field can be organized into four areas as shown in Fig. 1.
 
[Fig. 1] Major ICT utilization areas in the long-term care field

Below, we introduce ICT utilization needs and representative examples in each area.
(1) SNS
The long-term care service is carried out in a multidisciplinary collaborative play, and the enhancement of communication between staff members also affects the quality of the service. The "Medical Care Network (MCS)" provided by Japan Embrace is a private SNS service specializing in medical care and long-term care, and currently has 140,000 registered users. Many medical institutions are involved in home medical care and regional medical cooperation, but as an example specializing in long-term care, there are cases where long-term care providers who provide home-visit long-term care use it for communication and information sharing between staff.
On the other hand, many long-term care sites are lagging behind in the digitization of operations. On-site staff are still busy creating paper documents and diaries, taking over by writing notes in office notebooks, making inquiries and making reports to multiple facilities and related parties by telephone, and analog methods remain deeply rooted in the field. (Fig. 2).