AI applications in healthcare are becoming more common, especially to support doctors' diagnostic decisions and automate certain tasks.
However, artificial intelligence in medicine is an area that is not yet fully developed and can be considered not fully developed, even for automated robotic surgery arrangements that will certainly be updated.
Despite this, the large amount of data generated by artificial intelligence healthcare facilities; It contributes greatly to paving the way for predictive, more targeted and personalized health prevention policies, predicting requests for assistance, better and more accurate detection of symptoms, automated use of analysis results (images, laboratory analysis, etc.), formulating personalized care plans or protocols, analyzing detection of treatment side effects, predicting emergencies and calculating readmission rates to facilities, disseminating a pathology, and facilitating the coordination of care teams.
As technology and especially artificial intelligence enter many areas of life day by day, important developments are experienced in the field of health in this regard.
Artificial intelligence developments in medicine are desired to be widely used in both general health applications and drugs.
Artificial intelligence software and robots, which will assist doctors in many areas, allow healthcare services to be provided faster and more safely.
Artificial intelligence developments in medicine
Artificial intelligence causes significant revolutionary developments in the field of health as well as in all areas of life. Health services around the world also want to benefit from artificial intelligence.
Machine learning and assisted artificial intelligence have features that can develop an entire health system within the framework of a new vision.
According to the results of research on the subject, in the near future, everyone in the field of health, from specialist doctors to first aid workers, who are described as paramedics, will start to benefit from artificial intelligence technology.
According to estimates, the artificial intelligence market for the healthcare industry is expected to exceed $6.5 billion by 2021.
While 39 percent of the decision mechanisms in the health sector have investment plans for machine learning and predictive analysis systems, the investment amount is expected to increase significantly.
Artificial intelligence researches prepared with participants especially in Europe, Middle East and Africa regions reveal this situation.
Looking at the results within the scope of the research, the desire to use artificial intelligence and robots in the health sector is increasing day by day, and the main factor for this increase is shown as the ability of large segments to benefit from health services more easily.
A second finding is that diagnosis and treatment can be made much easier with the help of artificial intelligence. This situation shows itself as a factor that triggers the use of artificial intelligence and robots in the field of health.
In addition to these results, great importance is attached to the use of artificial intelligence in the health sector in order to ensure the spread of health services to wider areas. It is expected that the health services offered in various regions will increase, especially with the technological infrastructure.
Artificial intelligence to better guide patients
Since each patient can react differently to drugs and treatments, personalized treatment method stands out as an important area in the field of artificial intelligence.
Determining which treatment method to choose for patients normally takes time or is mostly performed by doctors by trial and error method.
Imagine listing your symptoms in an encyclopedia of all existing illnesses. This is the idea that Montreal CHUM is currently experimenting with for emergency triage.
Patients come to the emergency room, enter their information into a computer, which ranks them in order of urgency. AI also determines whether the problem is respiratory, pulmonary, cardiac, or other.
“We are currently comparing this machine-performed separation to human classification. The machine saves time, but we want to ensure that this triage is done intelligently and of high quality, because it may work well for one patient type but not another,” said President and CEO Dr. Fabrice Brunet CHUM says:
''You never believe that something will be useful just because it is new and innovative. We must remain critical. Like any innovation, AI must be evaluated and measured so we can deliver benefits.''
Better remote consultation with artificial intelligence
In hospital emergencies, as in triage, AI can be a valuable tool for guiding the patient remotely. Quebec telemedicine platform Dialogue implements AI that simplifies the care process.
“It essentially boils down to collecting a complete and accurate picture of the patient,” says Alexis Smirnov, CTO of Dialogue . For example, a patient with a skin problem explains to the chatbot Chloé the information, the symptoms and sends a photo of the problem.
The data and photo are then verified by a healthcare professional. If the next step involves making an appointment with a dermatologist, the process is again automated.
In this way, the doctor is asking the system to take the patient to the next step in their journey. The Dialogue team states that this tool will never replace doctors:
In Dialogue, they believe that AI technology is not advanced enough to make human, medical-based judgments, especially when you consider the human factors that come into play; types of decisions however said there is a big difference between making medical decisions and optimizing the non-medical components of the patient's care pathway.
Artificial intelligence to accelerate drug development
It takes tens of years and millions of dollars to get a drug to market. In epidemic cases such as Covid-19, the need for a pharmaceutical solution is urgent. One of the ways to reduce vaccine development time is by optimizing preclinical research.
With these in mind, InVivo AI , created by three Quebec doctoral students with the desire to speed up the drug development process , is being developed to serve this purpose.
Thus, it is aimed to emerge a technology that makes it possible to rationalize pharmaceutical research and development, their complementary expertise in molecular biology, computational neuroscience and machine learning.
“Currently, the drug development process is still pretty intuitive,” says Therence Bois, co-founder of InVivo AI ;
''For a specific therapeutic target, a researcher will often test a variety of molecules, quite randomly, and repeat the experiments until they find the active one for the target of interest in a single step. InVivo AI technologies analyze the data produced by these researchers and create models that simulate these experiments and a computational path to go through the process faster.''
Developing a diagnosis with AI
With the proliferation of medical devices, doctors are forced to account for more data. Today, the field of medicine where artificial intelligence is most prevalent is the interpretation of medical imaging and radiology.
Some cancers, such as lung or breast cancer, are very difficult to identify in images produced by scanners.
Programs can identify abnormalities that cannot be detected with the naked eye, thereby more reliably detecting early tumors and better targeting treatments.
Young Montrealer Imagia's mission is to accelerate the detection of certain types of cancer, develop new personalized treatments, and accelerate clinical research and development of new treatments.
The Evidens platform uses algorithms of a patented technology called Deep Radiomics to generate biomarkers from digital images.
These programs are capable of "self-learning" as they memorize all detected biological abnormalities , thus ensuring certainty in every diagnosis.
In this way, in-depth and personalized treatments for each patient are now more accessible.
AI can also help detect pathologies in highly sensitive places. Quebec company Diagnos has developed an AI that can detect diabetic retinopathy.
A complication of diabetes that affects 50 percent of type 2 patients and is responsible for 5 percent of blindness cases worldwide; It can detect the first signs of the disease from a photograph of the retina.
These photos are taken in a matter of minutes using special cameras available at various clinics, optometry centers and pharmacies.
The system has already analyzed the eyes of approximately 225,000 patients in 16 countries. André Larente, Head of Diagnos, says the system has managed to detect 98.5 percent of retinopathy cases.
Medical robots combined with artificial intelligence
Until recently, AI in healthcare was limited to research or predictive analytics. There is now a lot of focus on developing technologies that can improve robot-assisted surgery.
There are already very important uses that prove how artificial intelligence can improve techniques that have been used for several years in the field of robotic surgery, particularly microsurgery.
For example, researchers at Maastricht University Medical Center in the Netherlands used an AI-powered robot to microscopically stitch up tiny blood vessels, some as small as 0.03 millimeters.
The robot copied the surgeon's hand movements on a miniature scale.
Artificial intelligence was used to normalize these movements and prevent tremors or sudden movements from being transferred to the surgical robot.
Galen Robotics is working on similar technology for research-based otolaryngology procedures at Johns Hopkins University.
Some surgical robots perform a predetermined routine; they are not completely autonomous and surgeons still need to monitor operations and verify that procedures are performed correctly.
They can also interfere with the "editing" of the routine and improve its effects if necessary. However, scientists at Stanford University say that by 2050, artificial intelligence will enable fully autonomous surgical robots with only the control of the surgeon.
