Artificial intelligence is one of the trends shaping healthcare. It is relevant across developed countries, including Russia. These technologies are already being actively introduced and genuinely help both doctors and patients. In this article, we explore how AI is used in medicine, its applications, capabilities, and outlook.

How AI is used in healthcare

How does AI help in medicine? Artificial intelligence in healthcare covers many tasks, from routine, repetitive actions to complex work such as in-depth analysis of medical data and the discovery of hidden patterns.

The latest generations of neural networks process vast amounts of information in seconds. AI can detect conditions that even experienced doctors may not always see.

Introducing artificial intelligence into healthcare means using information networks, software, services, and devices. They make doctors’ work easier, speed up care, reduce diagnostic error risks, and develop treatment plans. Below, we consider AI applications across different medical fields.

Medical diagnostics

Diagnostics is a promising field for AI. Medical professionals can obtain an accurate, detailed picture of a patient’s health, enabling faster and more precise decisions. However, AI is an assistant rather than a qualified virtual healthcare professional. The specialist has the final say.

Artificial intelligence helps detect diseases through medical imaging, analyzing CT scans, MRI scans, and X-rays. Computer vision identifies abnormalities, including subtle tissue changes that the human eye may miss. AI also interprets these findings using an extensive knowledge base, offers a preliminary diagnosis, and provides information about it. Diagnostic accuracy reaches 94%.

Developing new medicines

AI’s development in medicine is reshaping modern pharmacology and changing how new drugs are created. AI can significantly shorten research timelines, assess medicine efficacy and safety, and determine treatment regimens.

How does it work? Machine learning algorithms analyze extensive biological material data and chemical compounds. This accelerates optimization of active substances’ molecular structures and selection of candidates for testing new medicines. AI also expands personalized medicine, including treatment of genetic diseases requiring an individual approach. It examines a person’s clinical picture, including DNA. The analysis produces recommendations for selecting or developing drugs and using them.

Here are several examples of successful AI use in development:

  • Comirnaty, a vaccine against coronavirus infection, developed by biotechnology company BioNTech together with manufacturer Pfizer.
  • Paxlovid, an antiviral medicine that helps combat COVID-19.
  • The Russian platform Syntelly, created to assess molecules’ toxicity and predict their activity.
  • Medicines for treating muscle fibrosis, released by Insilico Medicine in 2019. The company actively used artificial intelligence, which developed an algorithm in three weeks. Based on it, scientists selected candidates and tested active ingredients in animals in just 25 days. The best formula was identified in 46 days. Without AI, the search would have taken more than eight years.

Automating patient data management

Doctors previously had to complete medical records manually and spend considerable time finding them in an institution’s filing system. This made taking medical histories and diagnosis much more difficult and time-consuming.

AI assistants are also being introduced to help verify diagnoses against medical histories, quickly analyze patient data, and complete records.

AI assistants and chatbots for patients

Artificial intelligence in healthcare helps patients as well as medical professionals. Chatbots and digital online assistants become a first point of help, providing support around the clock.

They can:

  • book medical appointments;
  • assess symptoms from questionnaire responses;
  • remind patients to take their medication regularly;
  • track changes in health by analyzing key indicators such as pulse, glucose levels, and blood pressure;
  • assess well-being based on reported complaints;
  • answer health questions submitted as text or voice commands;
  • recommend nutrition, daily routines, and lifestyle changes;
  • provide psychological support through chat using cognitive behavioral therapy;
  • integrate with electronic health records.

Chatbots and AI assistants therefore help with diagnosis, treatment, prevention, and rehabilitation.

Automating doctors’ everyday tasks

Most countries now face an acute shortage of qualified doctors and mid-level healthcare staff. The WHO has published statistics and forecasts indicating that approximately 18 million workers will be needed for everyone worldwide to have access to healthcare by 2030.

The outlook for improving access to patient care is discouraging: societies continue to age, populations grow, and specialist medical education remains inaccessible to many. These factors increase demand for qualified doctors and make care harder to obtain.

AI frees medical staff from routine procedures, helps them complete electronic records quickly and accurately, analyze test results in detail, compile medical histories, and prescribe, monitor, and adjust treatment when needed.

Online consultations

Remote consultations matter in distant parts of the country. For some residents, an online consultation is the only way to obtain an expert opinion and receive help.

AI in medicine reduces patient service costs and makes telemedicine easier and more convenient. The technology can collect medical histories and support remote preliminary diagnosis. Some applications analyze symptoms, record consultations, and transcribe them. Google has also created an algorithm using retinal photography to detect diabetic retinopathy.

AI-powered medical applications

Many AI medical applications have been developed. Here are several examples:

  • Sensely is a conversational platform for interacting with a virtual assistant that checks symptoms, offers advice, assesses disease risks, and even forwards information to the treating doctor.
  • Ada assesses health through answers to specific questions. AI analyzes the information, identifies a possible problem, explores it in detail, and explains next steps such as lifestyle changes or examinations.
  • SberHealth was developed to build a clinical picture from symptoms, offer a preliminary diagnosis, and recommend further appointments with relevant specialists.
  • MedWhat is a virtual assistant that answers almost any question and analyzes X-rays, ECGs, and MRI results.
  • MedClueRx works in pharmacogenomics, studying how a person’s genes affect responses to medicines. After genetic testing, the service selects medicines and dosages and develops an individualized treatment plan.
  • SOPHiA GENETICS operates in more than 70 countries and serves over 790 medical institutions. Its main specialty is identifying genetic diseases and cancer.
  • Zebra Medical Vision analyzes imaging results for blood vessels, the heart, lungs, and other internal organs in detail.
  • Google DeepMind Health examines diagnostic data and symptoms and suggests diagnoses even for rare conditions.

AI adoption in Russian healthcare

AI is being successfully introduced and developed in Russia. Around 1,800 institutions across 72 Russian regions already use computer vision systems to analyze medical images. AI accelerates diagnosis by 50% and provides diagnostic accuracy of up to 87%.

Artificial intelligence remains experimental, but plans call for it to become everyday practice by 2030, creating a new, improved healthcare model. Important areas for adoption and development include AI-supported telemedicine, personal AI assistants, unified data ecosystems, and online diagnostics.

Services operating in Russia include:

  • TeleMD, designed for online consultations for cancer patients.
  • DOC+, a complete mobile clinic with information recording, doctor home visits, medicine reservations at online pharmacies, and quick smartphone access to electronic health records.
  • Voice2Med, a service that automatically converts speech to text in real time to simplify document completion.

Russia’s Internet Development Institute is also developing AI for independent diagnosis from MRI scans and X-rays.

The outlook for AI in healthcare

AI technologies have a promising future in medicine. Global leaders such as Apple, Google, and Microsoft are developing solutions in this field. Their work already helps make healthcare accessible to everyone, diagnoses accurate, and patient data organized.

Various services, applications, and platforms help detect early-stage cancer and other conditions, prevent disease, and support mental health.

In the future, AI can improve areas such as:

  • virtual nurses providing round-the-clock patient support;
  • surgery;
  • fast, accurate clinical trials;
  • administration;
  • detecting malicious activity, such as attempts to obtain confidential patient information from electronic databases;
  • preliminary diagnosis;
  • processing examination results;
  • developing treatment regimens;
  • monitoring treatment.

Potential challenges

Potential problems complicate AI’s prospects in medicine, including:

  • Privacy failures. Personal data may leak because systems are inadequately secured or information is shared with third parties without patient consent.
  • Limited reliability. Incorrect algorithm settings or insufficient data may cause errors.
  • Ethical concerns. If ethical standards are not built into an algorithm, AI may produce incorrect, unjustified, or even unlawful decisions.

Accountability. Users must understand that they cannot treat themselves with AI by taking every recommendation as an instruction. Application developers must take responsibility for possible consequences and clearly state from the outset that AI-assisted treatment without consulting actual medical professionals is unacceptable.

How to minimize risks

Improving healthcare with AI, evaluating its benefits, and preventing adverse consequences requires work in several areas:

  • Carefully prepare, structure, periodically review, and update neural network training data.
  • Provide additional medical staff training in AI technologies so doctors understand their capabilities and responsibilities when using these tools.
  • Protect data confidentiality through security measures: modern encryption, every patient’s consent to personal data processing, compliance with applicable laws, staff training in proper information handling, and regular AI algorithm testing.
  • Modernize the legal framework with input from several parties: ethics experts, medical staff, lawyers, and patients.

SkaiLab provides training, including corporate programs, consulting on organizational AI use, and integration and technical development of AI solutions for healthcare and other business sectors.

Training from SkaiLab includes:

  • A complete ecosystem for implementing AI technologies according to a plan tailored to your company, extending beyond an online course.
  • The potential to automate up to 80% of everyday business tasks, with ready-made AI scenarios for sales, analytics, marketing, reporting, financial management, and HR.
  • Practical benefits: topics with ready-made cases and products such as bots and agents, applying theory to develop real skills and tangible results that can be integrated into work.
  • Training and implementation in one cycle to close the last mile between knowledge and practical value: brainstorming sessions with case reviews, personal consultations, consulting, and ongoing support.
  • A structured, human, accessible format, without filler or online-course hype: live online classes, answers to every question, lesson recordings, and accessible materials covering all topics.
  • A practical collection of scenarios and tools: explaining complex AI concepts in accessible language without excessive jargon or irritating sales triggers, learning on current platforms, and moving from idea to code.
  • Exploration of all business processes with ROI assessment: classes divided by function, understanding key areas of company activity and their automation potential, and modules with measurable effects in leads, time, and savings.

References