In this blog post, we will examine the possibilities and limitations of artificial intelligence medical technology replacing doctors and, furthermore, leading humans to eternal life.
On October 5, 2011, Steve Jobs, an icon of innovative convergence, passed away. Though he created revolutionary products, he lost his life prematurely due to pancreatic cancer that was not detected early enough. The pancreatic neuroendocrine tumor he suffered from is an extremely rare disease, affecting fewer than one in 100,000 people annually. It is extremely difficult to detect the tumor visually using imaging devices. Therefore, diagnosis is based on the hormones released by the tumor and the symptoms it causes. However, in actual clinical practice, the disease is so rare that it is difficult to diagnose, and there are cases where no symptoms are present. Additionally, since diagnosis is impossible through blood tests, treatment is often delayed. However, in the future, even tumors that are difficult to diagnose early will be easily detected. A future where not only early diagnosis but also immediate treatment of discovered diseases is possible will arrive. The doctors who will give us early diagnosis and treatment in the future are artificial intelligence. In the future, various forms of medical artificial intelligence will be developed. Various medical artificial intelligence systems will collaborate with each other to form a capable team that will take charge of the diagnosis and treatment process. And this team will be able to go anywhere in the world in a very small form.
In Korea, there are no regular conferences on artificial intelligence for medical use. In contrast, in Europe, forums and conferences on AI in the medical field have been held regularly for over 30 years since 1985, and a specialized journal has been published since 1998. The conferences have focused on technologies that diagnose diseases and identify their causes based on data. AI excels at analyzing data compared to humans. Therefore, data becomes the core material for AI to demonstrate its capabilities. AI utilizes “big data” that has been meaningfully collected. Here, “big data” does not simply refer to a large amount of information. It refers to carefully refined information that brings new knowledge. Big data meets three conditions: rapid availability, vast quantity, and diverse forms of information. AI medical teams will be divided into three main tasks. First, a team is needed to collect data for disease diagnosis.
Next, there is a team that analyzes the obtained data to diagnose diseases. The data analysis and diagnosis team uses deep learning from electronic medical records to provide doctors with useful treatment methods and diagnoses. Once the diagnosis is complete, miniature robots play a role in enabling early treatment. Let’s first look at the team that obtains data from the AI-based medical team. To examine the role of the medical team in more detail, let’s assume a situation where images are captured to find cancer cells.
Here, the team is divided into two main groups. One group uses AI to improve the quality of the captured images, while the other uses AI to extract objective data. The AI responsible for improving image quality creates statistical models based on the data. These models determine whether the information in the images is useful for diagnosis. PGPD (Patch Group Prior Based Denoising) technology can be used in this process. PGPD technology removes noise—information deemed irrelevant to image interpretation—and replaces it with statistically modeled information derived from surrounding data. This process of identifying and correcting irrelevant parts of the image relies heavily on data. By leveraging pre-existing big data, AI can construct statistical models and apply them to image correction.
After correcting the image, data must be extracted from it. Before the development of artificial intelligence, there was no separate processing step after obtaining the captured image. Specialists would receive the photos as they were and use them for diagnosis. Recently, technology has been developed that allows computers to extract more objective information from captured images. The computer converts the image into digital information, analyzes the brightness of pixels and their context with surrounding areas, and quantifies this information into numerical values.
This method restores blurry and difficult-to-identify areas using the surrounding context. Artificial intelligence obtains information about the size, shape, and texture of the cells captured in the images. The information obtained about size, shape, and texture can be used as data to find cancer cells. Once the data extraction process is complete, a full analysis must be conducted to make a diagnosis. The diagnosis made by artificial intelligence is not simply based on the patient’s symptoms.
AI collects information based on accumulated big data to make a diagnosis. Big data is created through the process of extracting data, converting it into a single form, and storing it in an analyzable system. Artificial neural network models are used to utilize accumulated big data. An artificial neural network consists of an input layer, multiple hidden layers, and an output layer. The input layer first receives data and transmits it to the first hidden layer of the neural network with the highest degree of association. The data reaching the hidden layers undergoes complex mathematical processing to identify correlations and is then transmitted to other hidden layers before finally reaching the output layer. As the data moves through the hidden layers, it is processed, and the correlation with specific hidden layers is repeatedly adjusted to enhance accuracy. In a medical context, the input is symptoms and patient information, and the output is the diagnosis.
Once a diagnosis of the patient’s disease is made, it is ideal to treat the disease immediately. If the disease is not severe enough to require surgery, AI can be used for simple treatment. This involves injecting tiny robots into the bloodstream. These robots can detect the location of cancer cells, circulate through the bloodstream, and remove them. This method of injecting robots into the bloodstream is also highly beneficial for treating chronic diseases. Diabetes requires continuous blood sugar measurement and regulation. Therefore, artificial intelligence can constantly monitor the condition of the blood and determine when it is dangerous. If robots are present in the blood, hormones that regulate blood sugar can be used appropriately. In addition, artificial intelligence uses artificial neural networks to prevent medical errors during surgery. Quick decisions are required, and mistakes can be frequent, especially in critical procedures like coronary artery bypass surgery, where mistakes can be fatal. Artificial intelligence with a neural network model can quickly analyze the connection between decisions and situations to identify decisions that may be errors.
Once you understand the capabilities of an artificial intelligence medical team, you can imagine how they will play a role in the future. Going to the hospital is not easy or convenient. In the future, patients will not need to go to the hospital; instead, an artificial intelligence medical team will use a medical kit to treat illnesses at home. Medical kits could be distributed to every household. These kits can capture images of the human body or analyze blood samples. Using the captured images and blood samples, the kits utilize an AI system to extract data. This data is managed by a central control center, which stores individual family medical histories and medical records, enabling precise analysis of the images and blood data. The data is then analyzed through an artificial neural network model to deliver a final diagnosis.
The kit contains a device that can deploy miniature robots. The miniature robots receive the diagnosis and the individual’s biological information and perform appropriate treatment. If these medical kits are mass-produced and distributed, life expectancy will increase significantly in the future. Artificial intelligence technology is advancing at a rapid pace. Advancing artificial intelligence not only makes our lives more convenient but also extends our lives.
To extend human life, AI is being integrated into small kits that function as medical teams composed of multiple doctors. These medical teams perform three main roles quickly and efficiently: taking photos, enhancing video quality, and extracting objective numerical data from the enhanced videos that are difficult for humans to discern. A central control center stores big data on an individual’s family history and biometric information. Data from the central control center is sent to the AI system, which analyzes it by combining it with big data and extracted data. This analysis uses artificial neural network models. Based on the analysis, the AI system provides the final diagnosis. Even after the diagnosis, treatment can be administered immediately using ultra-small robots that enter the bloodstream. These robots, linked to AI, quickly eliminate the cause of diseases and enable treatment for chronic conditions.
In the future, nanotechnology will advance, and it will be possible to include multiple devices in a small space. Therefore, small-sized universal medical kits will be developed in the future. As explained earlier, universal medical kits will assist in the rapid and continuous treatment of human diseases with the help of artificial intelligence. Artificial intelligence will revolutionize the medical field by incorporating doctors into medical kits. In the future, hospitals may become unnecessary or even disappear. A society where diseases can be treated without traveling far or spending a lot of time is approaching. While extreme, as Ray Kurzweil has stated, the advancement of artificial intelligence technology could bring an era of human immortality closer than we imagine.
