Recently, we were surprised by the uproar caused by the "Chat GBT" chatbot, which relies on artificial intelligence to answer users' questions in a creative way, and there was widespread controversy on social media about its amazing ability to simulate human methods, which showed chatbots as if The maximum achievement of artificial intelligence in its history, but is this true?
In fact, if you want to know the real achievements of artificial intelligence, let go of that fuss, and let us tell you how “artificial intelligence” programs have revolutionized the fields of medicine, after solving one of the biggest mysteries of biology in modern times.
Incomprehensible protein
The DNA in the nucleus of human cells contains the information necessary for the formation of proteins in the body. It is like a genetic map containing about 20,000 proteins that together form the basic building blocks of living organisms.
In the year 2021, artificial intelligence techniques have been used to predict the structures of the vast majority of human body proteins.
This was a revolutionary event, given that understanding the shapes of proteins is extremely complex, difficult, and consequently costly, despite its great importance in developing and discovering new drugs, which may help treat many diseases for which we have been unable until today to find a suitable drug for them.
To reach these results, scientists used the new artificial intelligence program "Alphafold" to predict the structures of 58% of the amino acids that make up human proteins. In the end, the program was able to predict the structures of 350,000 proteins that make up the bodies of humans and other organisms, which is the most important contribution to intelligence. Synthetic in the development of medicine to this day.
"Artificial intelligence" programs such as "Alphafold", developed by the "Deep Mind" company, rely on "machine learning", as algorithms, mathematical models and data are used that make these programs able to learn in a human-like manner, and accordingly develop their own knowledge.
Discover a cure for liver cancer in 30 days
Artificial intelligence can speed up the discovery of new drugs, cutting down years of research.
(Shutterstock)
Predicting protein structures was not the only revolutionary work of the "Alphafold" artificial intelligence program. In January, the research journal "Chemical Science" published a landmark study that may herald the start of a new era in the world of pharmaceutical industries, after the program was able to The discovery of a new drug candidate for the treatment of hepatocellular carcinoma.
Hepatocellular carcinoma (HCC) is the most common type of liver cancer, and it ranks third on the list of cancer deaths, according to World Health Organization statistics in 2020, after it caused the death of 830,000 people around the world, a large number that is only surpassed by lung cancer deaths that exceeded one million. And rectal cancer deaths, which exceeded 900 thousand cases.
This study is the first practical application of the "Alphafold" artificial intelligence program in the pharmaceutical industry.
As the program was used to examine hepatocellular carcinoma to search for the weaknesses of those cells, which once discovered, another program works to determine the protein for treatment, and by predicting the structure of this protein through Alphafold, the process of generating therapeutic molecules for this protein becomes easy, it only needs to undergo tests biological, and then the stage of human experimentation.
The process of discovering and developing a new drug usually takes more than a decade, but the latest study shows how artificial intelligence can speed up the process of discovering new drugs and reduce years of research, after Alphafold was able to help develop an experimental treatment for hepatocellular carcinoma in just 30 days. .
Early detection of breast cancer
The researchers developed an artificial intelligence system with high accuracy.
It is able to detect breast cancer through ultrasound images.
(Shutterstock)
Hepatocellular carcinoma (HCC) was not the only malignant disease in which artificial intelligence showed promising results, as smart programs were recently used in the early detection of breast cancer, which has been known for many years to be difficult to predict, causing the suffering of thousands of women who started treatment after to late.
The beginning was when Regina Barzilay, a researcher in the field of artificial intelligence, was diagnosed with breast cancer in 2014, but the diagnosis came too late, which made her fear losing her life and wondering what would happen to her child next.
Regina was then dominated by a strong desire for many women to avoid that difficult experience, and she worked hard to develop an artificial intelligence-based system that could not only early detect breast cancer, but predict whether a patient was at risk of developing the disease within the next five years.
This promising technology is now able to save thousands of lives, after an artificial intelligence program was able to extract the unique patterns that are closely related to the disease.
This came after providing the program with the necessary data from patients' medical records and thousands of mammograms, so that it was able to understand the early diagnostic notes of the disease, and more than that, to predict patterns of the possibility of future infection.
In this context, the British magazine “Nature” published another study in 2021, heralding the ability of artificial intelligence to detect breast cancer using ultrasound, which was a thorny matter after ultrasound imaging devices were previously famous for giving false positive results regarding the detection of the disease. .
In this study, the researchers developed a high-accuracy AI system capable of detecting breast cancer through ultrasound images, and reduced false positive results rates by 37.3%, which highlighted the potential of future AI in efficiently detecting breast cancer and giving results. more accurate.
And psychiatry too
The Youper mental health platform works by using artificial intelligence techniques to reduce the symptoms of anxiety and depression in users, and helps them overcome those symptoms using their smartphones.
(communication Web-sites)
The matter does not stop at the limits of proteins, but extends to diagnosing the most complex diseases, such as mental illness, whose diagnosis requires a set of subjective and variable information, such as mood, sleep patterns, amount of daily activity, and other information that is difficult to accurately monitor and makes the diagnosis process depend in a large part. On the experience of the doctor.
The same difficulty a psychiatrist faces when predicting the likely outcome of an intervention, such as prescribing a particular medication or diet, or starting some psychological training.
This is what the consultant psychiatrist Alex Demetriou likened to predicting weather changes, pointing out that psychology depends mainly on knowing patterns, whether patterns of human behavior or the pattern of mental illness itself, and the ability to follow and monitor those patterns is what enables the doctor to make an accurate diagnosis of the disease.
Everything in our life depends on a certain pattern. The way we make morning coffee, for example, is a pattern, and the plan we make before starting a project also follows a pattern. Artificial intelligence in diagnosing mental illness.
Artificial intelligence programs rely on sorting large amounts of data and extracting recurring patterns from them. If these programs have the data needed to analyze patients’ moods, sleep patterns, and daily activity rate, they can make a huge leap in diagnosing mental illness.
Today, our mobile phones are filled with a lot of apps that collect data about us that we could not have imagined that we were producing, such as applications that track heart rate and counter the steps we walked during the day, and applications that track sleep patterns and moods.
All of this data can be used through artificial intelligence programs to extract our behavioral patterns that we may not be aware of, and it can also measure the effect of interventions, such as the result of trying a new drug or following a specific training, which may make these programs in the future an important tool in diagnosis and follow-up treatment. .
It does not stop at helping the psychiatrist in diagnosis, but artificial intelligence applications can encroach on the role of the doctor and practice his work in some treatment sessions.
As we have seen since the advent of the Youper mental health platform, which works using artificial intelligence techniques to reduce symptoms of anxiety and depression in users, and help them overcome those symptoms using their smartphones.
It seems like a normal therapy session, but it replaced the psychotherapist with an artificial intelligence program, and according to a clinical study at Stanford University published in July 2021, anxiety and depression rates among platform users have already decreased by rates ranging between 10-24% during two consecutive weeks of use, the matter Which made some wonder about the possibility of artificial intelligence replacing the psychiatrist in the future.
Improve care and reduce costs
Some rich countries have begun to use modern artificial intelligence technologies to improve the speed of disease diagnosis and to promote medical research and the drug industry.
(Shutterstock)
According to an article published by the Harvard School of Public Health "TH Chan", "artificial intelligence" technologies will enable health care professionals to address complex problems facing them, improve health care and reduce costs, as these programs are provided with a large database of patients' medical records and radiographs, The results of which can help doctors not only in diagnosis, but also in suggesting appropriate treatments, which has already been applied and achieved success in diagnosing diseases such as skin cancers and retinopathy for people with diabetes.
It's like the AI algorithms that a site like Facebook uses to recognize faces, or to suggest specific ads to you based on your recent searches.
In mid-2021, the World Health Organization published a report on the ethics needed to use artificial intelligence in the areas of health care, and the organization considered that smart programs have huge potential to improve the health of millions around the world if they are not misused.
Possibilities for misuse include the unethical collection and use of health data, as well as the potential for commercial interests of technology companies to control the rights and interests of patients.
Program designers and caregivers are currently working side by side, according to the report, to solve ethical dilemmas that they may face, in order to design and develop modern technology that takes into account ethics and human rights at every stage of development, and to preserve patient data from exploitation.
Some rich countries have already begun to use modern AI technologies to improve the speed of disease diagnosis and boost medical research and the drug industry.
But the report notes that artificial intelligence can fill the gap in medical care in poor countries that lack resources and whose patients do not have access to necessary medical care.