That’s correct.
Computer analysis of diagnostic images is one of the most common applications of AI in medicine. For example, in 2019, the neural network reviewed about 1,119 chest CT scans of patients from one of the clinics. It turned out that AI effectively identifies the risks of cardiovascular diseases.

Algorithms are able to detect signs of oncological diseases of the blood and bone marrow, lung and breast cancer, pneumonia, strokes, degenerative brain processes, glaucoma, diabetic and hypertensive retinopathy. Specialists from leading clinics teach the neural network what pathology looks like by placing it on hundreds of thousands of real images.

That’s incorrect.
Computer analysis of diagnostic images is one of the most common applications of AI in medicine. For example, in 2019, the neural network reviewed about 1,119 chest CT-scans of patients from one of the clinics. It turned out that AI effectively identifies the risks of cardiovascular diseases.

Algorithms are able to detect signs of oncological diseases of the blood and bone marrow, lung and breast cancer, pneumonia, strokes, degenerative brain processes, glaucoma, diabetic and hypertensive retinopathy. Specialists from leading clinics teach the neural network what pathology looks like by placing it on hundreds of thousands of real images.

That’s correct.
Such developments exist. For example, the company In silico Medicine is engaged in this. Using artificial intelligence and deep learning methods, it analyzes how different compounds affect cells, which compounds should be selected to achieve the desired effect and what side effects are possible. AI is used to search for new biomarkers and optimize candidate molecules, because on average, out of 10 thousand compounds, it shows a sufficient level of efficiency and safety and only one successfully reaches the market.

That’s incorrect.
Such developments exist. For example, the company Insilico Medicine is engaged in this. Using artificial intelligence and deep learning methods, it analyzes how different compounds affect cells, which compounds should be selected to achieve the desired effect and what side effects are possible. AI is used to search for new biomarkers and optimize candidate molecules, because on average, out of 10 thousand compounds, it shows a sufficient level of efficiency and safety and only one successfully reaches the market.

That’s incorrect.
One of the strengths of machine intelligence is the ability to process data in real time and immediately send it to specialists. This direction is actively developing. For example, some clinics already use a smart video analytics algorithm that processes the video stream coming from the ward or corridor of the hospital, and sends "casts" of events in the form of text alerts and screenshots to the nurse's remote control or smartphone. The system has been taught to notice dozens of events: for example, the module reports the risk of a fall, a prolonged absence of a patient in the ward, an attempt to get up from a weakened patient, and also signals the need to change the position of a patient with limited mobility to prevent the development of bedsores.

That’s correct.
One of the strengths of machine intelligence is the ability to process data in real time and immediately send it to specialists. This direction is actively developing. For example, some clinics already use a smart video analytics algorithm that processes the video stream coming from the ward or corridor of the hospital, and sends "casts" of events in the form of text alerts and screenshots to the nurse's remote control or smartphone. The system has been taught to notice dozens of events: for example, the module reports the risk of a fall, a prolonged absence of a patient in the ward, an attempt to get up from a weakened patient, and also signals the need to change the position of a patient with limited mobility to prevent the development of bedsores.

That’s correct.
Such systems were, in particular, tested during the coronavirus pandemic. One of the developments was created to automatically predict the risk of deterioration of the condition of patients with quid in the emergency departments of hospitals. The deep neural network studied radiographs of patients with positive PCR tests for Covid-19 and offered a prognosis of their condition for the next 24, 48, 72 and 96 hours. The system is configured to predict the potential use of a ventilator, referral to the intensive care unit or death of a patient. The development showed an accuracy comparable to the forecasts of experienced radiologists.

That’s incorrect.
Such systems were, in particular, tested during the coronavirus pandemic. One of the developments was created to automatically predict the risk of deterioration of the condition of patients with quid in the emergency departments of hospitals. The deep neural network studied radiographs of patients with positive PCR tests for Covid-19 and offered a prognosis of their condition for the next 24, 48, 72 and 96 hours. The system is configured to predict the potential use of a ventilator, referral to the intensive care unit or death of a patient. The development showed an accuracy comparable to the forecasts of experienced radiologists.

That’s incorrect.
The neural network is trained on photographs that were taken under certain conditions, usually in a doctor's office. And this means certain lighting, image quality, resolution and shooting angle. It is difficult to achieve the same quality with amateur photography. A neural network may mistakenly find signs of pathology where, in fact, a shadow has unsuccessfully fallen on the image. There are special applications for smartphones, but so far experts estimate their diagnostic potential is low.

Our NOTA mole tracker uses bioimpedance technology, that is, it measures real physical processes in real time, and does not "guess from a photo".

That’s correct.
The neural network is trained on photographs that were taken under certain conditions, usually in a doctor's office. And this means certain lighting, image quality, resolution and shooting angle. It is difficult to achieve the same quality with amateur photography. A neural network may mistakenly find signs of pathology where, in fact, a shadow has unsuccessfully fallen on the image. There are special applications for smartphones, but so far experts estimate their diagnostic potential is low.

Our NOTA mole tracker uses bioimpedance technology, that is, it measures real physical processes in real time, and does not "guess from a photo".

That’s incorrect.
Such developments are already actively used. Practice has shown that the proposed preliminary diagnosis is confirmed in about 70 percent of cases. The clinic doctor enters the patient's complaints into the service, and the assistant gives out the three most likely diagnoses.

It is necessary to make a reservation that the solution was created solely to help doctors, and the last word in the diagnosis is always left to the person.

That’s correct.
Such developments are already actively used. Practice has shown that the proposed preliminary diagnosis is confirmed in about 70 percent of cases. The clinic doctor enters the patient's complaints into the service, and the assistant gives out the three most likely diagnoses.

It is necessary to make a reservation that the solution was created solely to help doctors, and the last word in the diagnosis is always left to the person.

That’s incorrect.
The nature of eye movements can indeed indicate depression and other psychological disorders. At least there are studies on this topic, and the results are encouraging. However, this method has not yet been introduced into clinical practice.

That’s correct.
The nature of eye movements can indeed indicate depression and other psychological disorders. At least there are studies on this topic, and the results are encouraging. However, this method has not yet been introduced into clinical practice.


That’s incorrect.
Even a specialist may have problems developing a treatment strategy. In this case, he can ask the algorithm to suggest which similar cases of treatment existed. One of the problems that artificial intelligence can help solve is the treatment of patients on dialysis (with kidney disease).


That’s correct.
Even a specialist may have problems developing a treatment strategy. In this case, he can ask the algorithm to suggest which similar cases of treatment existed. One of the problems that artificial intelligence can help solve is the treatment of patients on dialysis (with kidney disease).

That’s incorrect.
Such operations have not yet been performed on humans. Although the Smart Tissue Autonomous Robot (STAR) robotic system has shown good results, having independently performed an operation to restore the integrity of the intestine. The machine connected the individual fragments of the intestine and stitched. According to developers from Johns Hopkins University (USA), the robot has already performed four similar operations, and all were completed successfully (data from January 2022). But clinical trials, that is, experiments involving humans, are still ahead.

That’s correct.
Such operations have not yet been performed on humans. Although the Smart Tissue Autonomous Robot (STAR) robotic system has shown good results, having independently performed an operation to restore the integrity of the intestine. The machine connected the individual fragments of the intestine and stitched. According to developers from Johns Hopkins University (USA), the robot has already performed four similar operations, and all were completed successfully (data from January 2022). But clinical trials, that is, experiments involving humans, are still ahead.