How will artificial intelligence revolutionise the future of healthcare and change our lives?

Using the example of Steve Jobs, we’ll illustrate the challenges of early diagnosis and treatment of rare diseases and explore the potential for artificial intelligence to drive future medical advances. The future of healthcare will be centred on AI’s diagnostic and therapeutic capabilities and the use of big data.

 

On 5 October 2011, Steve Jobs, an icon of innovation and convergence, passed away. His life was cut short by the failure to treat his pancreatic cancer early. His pancreatic neuroendocrine tumour is extremely rare, with fewer than one in 100,000 people diagnosed each year. It is very difficult to see the tumour visually using imaging modalities, so it is diagnosed by the hormones it releases and the symptoms it causes. However, in clinical practice, the disease is so rare that it”s difficult to diagnose, and in some cases, there are no symptoms. In addition, it is impossible to diagnose it with a blood test, so treatment is often delayed. However, in the future, it will be easier to detect tumours that are difficult to diagnose early. Not only will we be able to diagnose them early, but we will also be able to treat them immediately. The doctor of the future who will give us early diagnosis and treatment is artificial intelligence. In the future, various forms of medical AI will be developed. They will be able to work together to form a competent team that will be in charge of the diagnosis and treatment process. And this team can be located anywhere in the world in a very small form.
In Korea, there are no regular conferences on AI in healthcare. In Europe, on the other hand, forums and conferences on AI in healthcare have been held regularly for more than 30 years since 1985, and a specialised journal has been published since 1998. The main focus of these conferences has been on technologies that use data to diagnose diseases and find their causes. AI is better at analysing data than humans, so data is the key ingredient for AI to perform. AI makes use of meaningful collections of “big data”. By “big data,” we don’t just mean large amounts of information. It refers to information that has been refined to bring new knowledge. Big data fulfils three conditions: it is quickly available, vast in volume, and diverse in form. An AI medical team will have three main tasks. First, the team will need to collect data to diagnose the disease. Then there is a team that analyses the data to diagnose the disease. The data analysis and diagnosis team will use deep learning from electronic medical records to provide doctors with useful treatment methods and diagnoses. Once the diagnosis is made, tiny robots can be used for early treatment.
Let’s take a look at the data-gathering part of the AI healthcare team first. To see the role of the medical team in action, let’s consider a situation where an image is taken to find cancer cells. The team consists of two main parts There is an A.I. that improves the quality of the images and an A.I. that obtains objective data. The AI that improves the image quality creates a statistical model based on the data. Based on the statistical model, it determines whether the information in the image provides useful data for diagnosis. Patch Group Prior Based Denoising (PGPD) technology can be used here. PGPD technology removes the noise, the information that is deemed unhelpful for image interpretation, and replaces it with new information that has been compared to the surrounding information and organised by the statistical model. Data is key in the process of finding and correcting unhelpful parts of the image. Using big data, AI can build statistical models and use them to correct the footage.
Once the footage has been corrected, data needs to be extracted from it. Before the development of artificial intelligence, there was no processing involved once the images were taken. A specialist would take the images as they were and use them for diagnosis. More recently, technology has been developed so that computers can extract more objective information from the images. The computer converts the image into digital information and then analyses the brightness of the pixels and their context with the surroundings to produce a quantitative value. It reconstructs blurry and hard-to-see areas using the surrounding context. The AI obtains information about the size, shape, and texture of the cells in the image. This information about size, shape, and texture can then be used to detect cancer cells.
Once the data extraction process is complete, it needs to be analysed to make a diagnosis. The diagnosis made by A.I. is not just a simple knowledge based on the patient’s symptoms. It 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 form, and storing it in an analysable system. Artificial neural network models are used as a way to use accumulated big data. An artificial neural network consists of an input layer, multiple hidden layers, and an output layer. The input layer receives the data first and passes it to the first hidden layer of the neural network with the greatest degree of association. Once the data reaches the hidden layer, it is mathematically processed in a complex way and passed on to other hidden layers in search of associations before finally reaching the output layer. At each hidden layer, the data is processed and the degree of relevance to a particular hidden layer is iteratively modified to ensure high accuracy. In a medical context, the input is the symptoms and patient information, and the output is the diagnosis.
Once we have a diagnosis of the patient’s disease, we want to be able to treat the disease immediately. If the disease does not require surgery, it can be treated in a simple way using AI. This is done by injecting tiny robots into the blood. These tiny robots are able to recognise the location of cancer cells and can travel through the blood to locate and eliminate them. The use of robots in the blood can also be very helpful in the treatment of chronic diseases. Diabetes requires constant measurement and control of blood sugar. Therefore, an artificial intelligence can monitor the condition of the blood at all times and identify dangerous conditions. If there”s a robot in the blood, the hormones that regulate blood sugar can be used appropriately. In addition, AI uses artificial neural networks to prevent medical errors during surgeries for treatment. It can prevent mistakes in coronary artery bypass grafting procedures, which require quick judgement and can be prone to mistakes. By building neural network models, AI can analyse the relevance of a decision to the situation in a short period of time, and catch suspected mistakes.
Now that you know what AI medical teams are capable of, you can imagine them in action in the future. Going to the doctor is not easy and difficult. In the future, patients won’t have to go to a hospital, and AI healthcare teams will be able to treat illnesses at home using medical kits. Medical kits can be distributed to every home. The kits can take pictures of a person’s body or analyse their blood. Through the photos and blood, the kit uses the AI system and obtains data. The data is managed by a central control centre. The centre stores the individual’s family history and medical history so that it can precisely analyse the photo and blood data. The data is analysed using an artificial neural network model and a final diagnosis is made. The kit contains a device that allows the insertion of a miniature robot. The robot receives the diagnosis and the individual’s biometric information and performs the appropriate treatment. If these medical kits are produced and distributed in large numbers, they will greatly increase our lifespan in the future.
Artificial intelligence technology is advancing at a furious pace. It not only makes our lives easier, but it can also extend our lives. In order to extend human life, AI fits into a very small kit and creates a medical team of doctors. The medical team can quickly and easily fulfil three main roles. It can take pictures, and it can correct the images to make them better, not worse. They enhance the images and then derive objective numbers from the images that are difficult for humans to judge. The central control centre stores big data about an individual’s family history and biometric information. The data obtained from the central control centre is sent to the analytics platform, which synthesises the big data and the extracted data. An artificial neural network model is used for analysis. The final diagnosis is made through data analysis. Even after the diagnosis is made, a small robot that enters the human blood can be used for immediate treatment. In conjunction with artificial intelligence, tiny robots can quickly eliminate the cause of the disease and treat chronic diseases.
In the future, nanotechnology will allow for the inclusion of multiple devices in a small space, which will allow for the development of small, all-round medical kits. These kits will help humans quickly and continuously treat diseases with the help of artificial intelligence, as described earlier. AI will disrupt the medical field by putting a doctor in a medical kit. In the future, hospitals will be unnecessary in the long run and may even disappear. A society where you don’t have to travel far or spend a lot of time to treat your illness is on the horizon. In an extreme case, as Ray Kurzweil suggests, advances in artificial intelligence could make it possible for humans to live forever.