Patients will soon be able to rely more on artificial intelligence than on humans

Artificial intelligence continues to hold promise in improving medical care.

For example, doctors at the Mount Sinai Health System in New York City used AI monitoring patients in their ‘step down’ units. These are patients who are not quite sick enough to require intensive care admission, but whose condition can deteriorate rapidly with minimal warning. The AI ​​systems monitored the patient’s vital signs, heart rhythm, laboratory results, and nurse observations. The patients were divided into two groups: those monitored with AI versus those monitored with traditional methods. If the AI ​​detected the likelihood of clinical deterioration in patients in the first group, it sent an alert to the rapid response medical team to recommend administering appropriate therapy.

The researchers found that patients whose vital signs were monitored with the AI ​​were 43% more likely to receive medications to support the heart and circulatory system compared to patients monitored with traditional methods. In addition, the patients followed with AI had a lower mortality rate at 30 days (7%) compared to the group followed with traditional methods (9.3%).

Senior study author Dr. David Reich observes“We view these as augmented intelligence tools that accelerate personal clinical assessments by our doctors and nurses and drive the treatments that keep our patients safer. These are important steps towards the goal of becoming a learning healthcare system.”

Another team of researchers assessed the ability of chatbots such as ChatGPT-3.5 and ChatGPT-4 answer specialized medical questions such as “How should you manage a patient with known cirrhosis who presents with new ascites?”

The answers were assessed by eight doctors, including specialists in the relevant fields. They found that “both ChatGPT models received high marks for accuracy, relevance, clarity, benefit, and completeness. However, GPT-4 scored higher on all criteria.” Furthermore, ChatGPT’s power lies in its ability to quickly access a wide range of medical data from different sources. By providing physicians with immediate access to the latest findings, clinical standards and specific cases, ChatGPT acts as a catalyst to keep them aligned with the ever-changing medical landscape. This ability enhances physicians’ ability to make informed judgments when dealing with complex or unusual medical scenarios.”

Given these dramatic results, many patients are interested in AI-augmented healthcare. According to one survey, “64% of respondents said they would trust an AI diagnosis over a human doctor. This percentage is growing even more among Generation Z, with four in five of this generation claiming to trust AI over a doctor.”

I don’t think AI is anywhere near ready to replace human doctors. Recently, Google’s AI overview was criticized offering hilariously bad medical advice to patient questions. For example, when the Google AI was asked, “How many bricks should I eat?” it recommended to eat “at least one small stone per dayand suggests hiding “loose rocks in foods like peanut butter and ice cream.” The incorrect answer was apparently partly taken from a satirical article The onion.

But in the right hands, AI can certainly augment human doctors, who can’t always keep up with all the nuances of the latest literature. AI systems will not be limited by the need to sleep, eat, or attend to their personal lives. Many physicians already rely on “physician extenders,” such as nurses or physician assistants, to help with busy workloads. I can easily imagine a day in the near future when AI will be yet another form of physician extension – perhaps even more reliable than humans.