AI bot Kredo Telegram channel
Please leave your request and we will contact you
Required field
Required field
Неверный ввод
Согласие обязательно
Написать в WhatsApp Проверить досье в Telegram-боте
Treatment Selected Before the Doctor's Visit: AI in Medicine
07.04.2026

Treatment Selected Before the Doctor's Visit: AI in Medicine

Selecting treatment for a patient before they even see a doctor is a practice that is gradually becoming the norm. As Yegor Kashka, COO of KREDO.TECH, explained, these capabilities have emerged thanks to big data and artificial intelligence: systems analyze millions of cases and identify patterns that no specialist could process manually. But does this mean doctors will soon be replaced? The expert shares his view.

Source: 1.ru publication

Selecting a patient's treatment before they even see a doctor is a practice that's gradually becoming the norm. As Yegor Kashka, COO of KREDO.TECH, explained, these capabilities emerged thanks to big data and artificial intelligence: systems analyze millions of cases and suggest treatment options in advance. As a result, medicine is moving away from the "trial and error" method and increasingly relying on precise calculations tailored to a specific patient.

How the patient's "digital footprint" came about

Just ten years ago, medical information was scattered. Some was kept in paper archives, some in different clinics, some in separate files. Bringing it all together was nearly impossible.

This fragmentation long remained one of the main barriers to developing a personalized approach in medicine.

Today the situation has changed. Virtually all medical organizations in the country are connected to a unified healthcare information system, where hundreds of millions of records are accumulated. This is a "living" database that is constantly updated and used for analysis.

The expert explained that this volume of data matters especially in complex cases.

Where AI is already genuinely at work

The clearest example is radiology. A radiologist reviews dozens of scans in a shift and can miss barely noticeable changes. An algorithm, however, doesn't get tired, since it compares a new image against millions of previous ones and highlights suspicious areas.

In Moscow in 2025, AI-based systems helped process more than 14.4 million radiology studies, and by early 2026, more than 17 million.

Today, such services are used across 43 clinical areas and help detect signs of a wide range of diseases — from cancer to osteoporosis.

The specialist stressed that the algorithm doesn't make a diagnosis — it works as a "second set of eyes." The final decision remains with the doctor.

What's changing for doctors and patients because of technology

Previously, a doctor had to gather information almost by hand: test results might be in one system, scans in another, and discharge summaries in a third.

Technology now makes it possible to pull data into a single window, where the system generates a preliminary picture of the patient's condition in advance. This speeds up decision-making, reduces the burden on doctors, lowers the chance of errors, and reduces the risk of burnout among specialists, the expert believes.

The effect of these changes is also visible in practice. For example, preliminary annotation of CT scans cuts interpretation time by about 30%.

Genomics makes treatment individual

Genetics has become a distinct branch of personalized medicine. Not long ago, full genome sequencing was an expensive, lengthy procedure available to a limited number of patients, but today the situation has changed markedly.

By October 2025, Russia had already built a database of 100,000 sequenced genomes. The cost of analysis has fallen at the same time: it now averages around 80,000-100,000 rubles, depending on the study's parameters.

Will AI replace the doctor

As technology develops, the question of whether AI can replace the doctor comes up more and more often. In practice, though, this isn't about replacement — it's about redistributing tasks, Yegor Kashka explained.

Algorithms take on routine operations, including data retrieval, image analysis, and generating preliminary conclusions, while the doctor focuses on decision-making and interacting with the patient.

Yegor Kashka, COO of KREDO LLC

Frequently asked questions

How can AI suggest a treatment before a patient even sees a doctor?
Systems analyse millions of cases using big data and artificial intelligence and suggest treatment options in advance, so medicine moves away from "trial and error" and relies on precise calculations tailored to a specific patient.
Does the algorithm replace the radiologist?
No. The algorithm does not make a diagnosis — it works as a "second set of eyes", and the final decision remains with the doctor.
How widely is AI already used in radiology in Moscow?
In Moscow in 2025, AI-based systems helped process more than 14.4 million radiology studies, and by early 2026 more than 17 million. Such services are used across 43 clinical areas and help detect signs of diseases from cancer to osteoporosis.
How much time does preliminary annotation of CT scans save?
Preliminary annotation of CT scans cuts interpretation time by about 30%.
How much does genome analysis cost now?
The cost averages around 80,000–100,000 rubles, depending on the study's parameters. By October 2025, Russia had already built a database of 100,000 sequenced genomes.

See also

Expert insights on medical device registration
Regulatory updates, case studies and checklists — in our Telegram channel.
Subscribe