[Music] Bakkumi TV Bakkumi [Music] TV Hello, I'm Park Geum-mi. Recently, with the continuous advancement of AI technology, it is expanding into various industries. Among them, AI technology is showing amazing synergy in the medical field. Today, to share related knowledge with us, we have invited a guest. Hello, could you please introduce yourself? Hello, everyone. Hello, I'm Kim Jae-woong, an employee of Kiumje. Today's topic is Medical AI. It's well known that AI is amazing, but the anticipation is high when AI technology is combined with life-saving medicine. AI technology is showing good synergy with the medical industry, so you can look forward to interesting content. Not long ago, earlier this year, the JP Morgan Healthcare Conference was held in San Francisco, USA.

The turning point was NVIDIA's keynote speech. It was about using AI technology for drug development. Now, has AI technology come this far in making medicines already? Earlier this year, there was the JP Morgan Healthcare Conference. This event includes the world's largest global investment symposium in the field of bio-medicine. Here, NVIDIA actually unveiled a platform for drug development called BioNemo that utilizes AI technology. NVIDIA CEO Jensen Huang mentioned that it can dramatically shorten the existing drug development period, process existing data seven times faster, and save costs significantly. By mentioning this, it has attracted a lot of people's attention. However, although we know that AI technology is impressive, I wonder if it has been applied in so many places. How is it used in medicine?

Has the partial application been applied well? As you mentioned, after the emergence of generative AI technology, AI technology continues to be applied in various industries. Among them, especially in terms of improving productivity in industries, we can point to the medical field. The medical sector itself requires a long learning process for each patient's different disease patterns and treatment procedures, as it is an area that requires expertise and accumulated experience. This is all about the process of accumulating a large amount of data. Therefore, the AI technology that processes a large amount of data in a short time to derive results is a very conducive environment for enhancing productivity. I think I can say this. Oh, in the drug development process, it is essential to extract a lot of data from many patients, process this collected data well, and derive meaningful results.

AI is good at that, so it seems to be well applied in the medical field. Then, more specifically, what is medical AI, how is it being used? Medical AI refers to software that supports patients and medical staff by going through various processes such as machine learning, deep learning, and natural language processing to understand various healthcare information. There are roughly four main areas where it is applied. First, in the diagnostic field. When we go to the hospital, we take X-rays, CT scans, MRIs, and so on. By training AI with various image information, doctors can diagnose by quantifying, for example, the probability of cancer in a specific location based on the image information. Secondly, for patients, monitoring vital signs such as heart rate, respiration, and body temperature continuously and providing real-time information on the probability of conditions like cardiac arrest through scoring. This is how AI directly diagnoses.

It plays a role in increasing the probability of diagnosis by assisting the doctor from the side, rather than the concept of making decisions. Doctors are human, so they can get tired. If they are tired, they may make wrong diagnoses, and they cannot check the patient's pulse, breathing, body temperature, etc., 24 hours a day. AI does not get tired and does not sleep. Therefore, it can catch the parts that humans can miss, and the probability of diagnosis increases. It would be good to think like this. It is not that AI diagnoses, but that AI helps humans diagnose. This is the concept. Then, what about the second one? Secondly, in terms of drug development, through simulations, candidate substances are derived, and the prognosis of how patients will react to the drug is predicted. AI technology is being used in this process. Drug development typically takes more than 10 years.

It is a large-scale project that takes time and costs an average of about 2.5 billion dollars, so typically, major global pharmaceutical companies have focused on developing new drugs. I think I should explain a bit about the process of developing new drugs. First, they select diseases with therapeutic potential and targets, then go through the process of creating new substances that can expect therapeutic effects or optimizing existing structures to select drug candidates. This process usually takes about 5 years. After that, as you know, it goes through clinical trials before being sold. Each clinical trial period can take from 1 to 2 years, up to a maximum of 5 years. Even then, there are numerous cases where it fails by the time it reaches phase 3. That's why drug development is likened to a camel passing through the eye of a needle. Recently, AI technology has been utilized in this process.

It is giving the effect of reducing the time and development costs. For example, during the Corona pandemic, vaccines were supplied faster than expected. In fact, when developing a treatment for COVID-19, AI supercomputers were used. It is said that a task that used to take over a year to identify candidate substances was done in just ten days. When AI is applied to drug development, which can be called the flower of the bio industry, such results occur. No wonder vaccines came out so quickly during the pandemic. Although experts said it was impossible, the secret was AI. Then, the third one is genetic analysis. Since people have their own genetic traits, a lot of genetic information is needed to realize precision medicine. In this process, when analyzing the characteristics of proteins, AI can increase the processing capacity.

It is said that the fourth aspect is the support in medical record data. Various clinical papers make medical record information easy to read by presenting it in a report format or utilize AI voice recognition technology to process and interpret conversations between patients and medical staff in natural language, providing recommendations based on that. Medical AI can efficiently and, above all, quickly analyze vast amounts of medical data, thereby preventing medical staff errors, increasing efficiency in medical decision-making, and even leading to cost savings. It has already come this far, with AI entering the medical market, aiding in diagnosis and drug development. Ultimately, from an investment perspective, one may wonder how much growth potential this can continuously have. The growth potential of medical AI industry, how is it? Let me explain the growth factors of the medical AI industry in three aspects.

First of all, I can say that the medical AI market is expected to grow at a high rate of around 40% on average in the future. The global medical AI market is expected to continue its high growth at an average rate of nearly 37% from 2021 to 2030, forming a market of approximately $190 billion, roughly 260 trillion won, by 2030. According to the Korea Institute of Science and Technology Information, in the case of South Korea, the market size was about 77.3 billion won in 2020, and it is expected to continue high growth at an average rate of 46%, forming a market size of 745 billion won by 2026. By region, the United States is expected to account for approximately 40% of the total market, and the Asia-Pacific region, including South Korea, is forecasted to form a market of nearly 180 billion dollars, roughly 25 trillion won, similar to Europe in 2027. In terms of applications, in the field of medical imaging diagnosis, in 2027.

It seems that the market will be formed most significantly with about 122 billion dollars, growing at an annual rate of 117% in the drug discovery field, which is expected to create a market of 16 trillion dollars. An annual growth rate of 40% is truly an extraordinary figure. The field that seems to grow the most significantly is diagnostics, and it has been summarized that drug development is the area expected to grow the most significantly. Next, the second point is about being able to grow structurally. It is promising in terms of helping to solve the medical shortage problem. As you know, global aging is progressing, and the elderly population is increasing rapidly. Since medical expertise is essential and the supply of medical personnel is relatively limited, there have been issues such as insufficient beds and medical staff shortage in the past.

This is a situation where there is a forecast of a shortage of medical personnel structurally. The problem is that when doctors become more fatigued, it eventually leads to medical accidents due to misdiagnosis. According to the Korea Consumer Agency, from 2017 to 2021, in terms of the causes of cancer misdiagnosis in South Korea, 31% was due to errors in image interpretation, and 40% of the additional tests due to interpretation errors accounted for the largest proportion. In reality, when introducing medical image interpretation solutions utilizing AI, you can see a much higher success rate in diagnosis. In other words, in the future, there will be no need to search for renowned experts by region. This is how you should think about it. Thirdly, the medical AI industry is a sector that is being supported by governments worldwide. In line with South Korea's government policy direction for fostering digital healthcare, new medical technologies are being introduced.

The system is continuously being improved. In the case of South Korea, new medical technologies are based on a system of pre-market approval after evaluation. It is a system where technologies with confirmed safety enter the market first and then accumulate clinical data. The South Korean government is expanding the scope of advanced new medical technologies and increasing the number of institutions, lowering barriers to market entry to help grow the overall medical AI market. Not only in South Korea but also in advanced countries, regulations are being continuously relaxed for the growth of the medical AI industry. For example, in the United States, through the relaxation of regulations on medical AI, FDA approval for AI medical devices is accelerating. In 2021, about 115 AI-based medical devices received FDA approval, and as of last October, it has been reported that over 500 FDA approvals have been accumulated. So, how should one go about concentrating investment in medical AI?

It is a bit different from what we already know. In the existing biotech sector, which is mostly dominated by pharmaceutical companies, medical device companies account for a small portion. However, the medical AI sector is somewhat different. Medical AI companies target existing companies in the biotech and medical device sectors to provide AI solutions such as diagnostics, new drug development, and analysis. They utilize AI technology to develop new drugs or provide programs that assist in diagnosis, as I mentioned earlier. Therefore, it can be said that they are different from traditional pharmaceutical or medical device manufacturers. In other words, they are companies that supply technologies that enhance productivity in the existing healthcare industry and have the characteristics of the IT industry.

There are many companies, and this can be considered a distinguishing factor. That's right. In the end, medical support is ultimately a technology, so companies with such technology are more involved here, like tech companies. This can be seen more. So, in the end, are companies like NVIDIA included in this ETF? Yes, Microsoft, Oracle, and NVIDIA, big tech companies, have been continuously expanding their partnerships in medical AI investments recently. Google, along with OpenAI, is said to have AI that can pass the US medical licensing exam, so big tech companies are also showing a lot of interest in the medical industry. I think I can say this. However, in the case of medical AI, the difficulty level from development to commercialization is very high. Medical data such as genetic information is mostly unstructured data, so it is challenging to apply AI to it.

It is difficult to teach intuitively, and large language models like Chat GPT are currently dominating the field of generative AI, making it challenging for more intuitive learning methods to take hold. Commercializing AI in areas such as medical diagnosis and drug development may take some time. Additionally, obtaining a large amount of data in sensitive areas like medical information is difficult due to the sensitivity of personal information. Therefore, in the medical AI industry, it can be seen that there are relatively high entry barriers for existing big tech companies. For these reasons, big tech companies in the medical AI field are showing a somewhat passive attitude. Instead of actively acquiring medical AI companies, they are expanding the application of their original AI models to the medical industry. This means that research and development are progressing by expanding the application scope of the original AI models in the medical field.

It focuses on providing services such as medical information solutions and electronic health records, rather than directly engaging in diagnosis or drug development, even if it acquires a medical AI company. It could be said that rather than investing in companies, Korean companies are more specialized in the medical AI industry. There are many good companies in Korea as well. In the medical AI market, it seems that Korean companies have some competitiveness. In the case of Korean medical AI companies, most of them have developed AI specifically for the healthcare field from the beginning, rather than applying big tech AI models to medicine. For example, in the case of diagnostic AI, they first started learning data based on medical images such as X-rays and MRIs, and a lot of consideration has been given to how this can actually be used in medical settings.

Many AI medical diagnoses are structured to be very user-friendly for medical professionals to use. For example, a well-known company is Lunit, which has two AI services: Lunit Insight, a cancer diagnosis imaging biomarker solution, and Lunit Scope. Lunit Insight is currently being used in 1,500 medical institutions in 40 countries worldwide. Another company, VUNO, provides a solution called DeepCARS, which predicts cardiac arrest. DeepCARS has been implemented in over 80 hospitals, and notably, no hospital that implemented it has reported any incidents. It is expected that over 120 hospitals will adopt DeepCARS this year, with the possibility of FDA approval by the third quarter of this year.

It is a promising situation. Among the AI pharmaceutical development companies, there is a company called Voronoi AGO. Voronoi specializes in precision-targeted therapy design and development, particularly in developing cancer treatments. They have their own laboratory that can extract candidate substances in the shortest time by combining AI, called the Voronoi Mix platform. Based on this drug design ability, they have shortened the development period of candidate substances from 3 to 5 years to about 1 to 1.5 years. They transfer these candidate substances to other bio-companies through the conventional method of technology transfer. They have conducted five contracts so far, achieving around 3 trillion won in scale. Companies are doing well in this regard, and their technological capabilities are also relatively high compared to major global countries. The technological gap between the US and South Korea in AI diagnostic and treatment systems is quite significant.

When compared to 2016, the technology gap in AI disease prevention and prediction has decreased from 3.5 years to 2.9 years. If we compare this with the widening technology gap between Europe and Japan during the same period, we can see that South Korea's medical AI technology has grown even more. Additionally, when looking at the number of clinical AI research and development projects by country during the same period, South Korea ranks third after China and the United States. The medical market is a leading industry of the future, and with the integration of AI, it is showing significant growth in this vast market. An ETF that allows investment in such an attractive industry seems to have been introduced today, named the Kosep Medical AI ETF. Could you tell us more about this ETF? Our Kosep Medical AI ETF includes up to 25 domestic companies that provide AI diagnostics, AI drug development, and medical AI platform services, weighted by market capitalization.

There is an ETF that invests in the medical AI cancer diagnosis company I mentioned earlier, Lunit, as the single largest component. In addition, companies like Vieworks and Celltrion Healthcare, as I mentioned earlier, are included in the top 10 components. It specializes in investing in companies that provide medical services using AI technology, rather than traditional healthcare companies. This can be considered a characteristic. The medical AI industry is in the early stages of growth, as I mentioned before, and it is a very appropriate investment vehicle for the expected high long-term growth of the medical AI industry. In the short term, there is an expected increase due to the profitability of AI diagnostic companies reaching a breakeven point by the end of this year and next year. In the long term, we can also expect an increase based on the approval of AI new drug development.

Also, many stocks rose a lot last year and are currently experiencing significant corrections. Therefore, it can be said that it is not a bad situation from an investment timing perspective. Can I buy in my pension account? Yes, you can buy in your pension account. What is the fee? It is 0.49% per year. Even though this is an advertisement video, we have prepared a big comment event for our subscribers who watch the video until the end. If you leave a message of support in the comments right now, you will automatically participate in the event. It doesn't matter who you support, whether it's supporting KOSPI, supporting my channel, or supporting the health of people around the world. Among those who leave comments, we will randomly select 30 people and prepare Baskin-Robbins variety packs as prizes. Since the weather has gotten hotter, please enjoy some refreshing ice cream. We have also prepared a big surprise.

I will select one person who left a thoughtful comment and this time, it's Mr. Kim. I will send a mini gold bar card packing type, 1.875g, as a gift. Exactly one week later at 9 p.m., I will announce the winner through the community. We have prepared abundant event prizes, so I hope all of you, our subscribers, will participate and receive a lot. Today, I introduced the ETF through the advertising video. It is Kosep Medical AI ETF. I hope it will help with your medical AI investments. This is the end of today's content. Thank you. Thank you. Park Kkomi TV, Bakkumi TV.

[광고] 🩺 AI가 하다하다 이제 사람까지 살릴기세 l KOSEF의료AI ETFㅣ키움투자자산운용ㅣ푸짐한 댓글이벤트도 있어요🎁

*채널목표*
박곰희TV는 "내 돈은 내 손으로" 라는 채널 슬로건처럼,
많은 사람들이 스스로 자산관리 하는 방법을 익히는 것을 목표로 하고 있습니다.

*기억해주세요*
박곰희TV는 '종목추천'이나 '주식리딩'을 하지 않습니다.
박곰희TV는 카카오톡방, 텔레방 등을 일체 운영하지 않습니다. 모두 사칭방이니 유의하시기 바랍니다.
박곰희TV는 소속 금융사가 없습니다.
박곰희TV에서 진행하는 광고성 컨텐츠는 모두 '광고임'을 표기합니다.

*비즈니스 문의*
gomhee.park@gmail.com
