When you fall ill and need medical attention, would you trust a clinic entirely run by Artificial Intelligence (AI), or would you feel more comfortable with a human doctor making the call?
Dr Keren Priyadarshini, Regional Business Lead, Worldwide Health, Microsoft Asia, offers a third choice: One that is run by both.
In healthcare, AI serves as a “second reader”, she explains. “The technology cannot diagnose or give medical advice because you need someone skilled to validate and make that final call.” Instead, AI operates more like an untiring assistant, constantly alert, guiding the doctor with suggestions like, “Hey, this is probably what you should be looking at.”
This symbiotic relationship between AI and healthcare professionals highlights a new era in patient care; one where technology complements, rather than replaces, human expertise.
In hospitals, for example, AI helps to streamline medical reports and enable clinical intelligence by making sense of vast amounts of data. Patients benefit from improved health and treatment outcomes, as well as a better quality of life.
Then there’s life sciences, where the impact of AI on drug development is especially promising. “In the past, a drug would take about 10 to 20 years to go to market,” says Dr. Priyadarshini. “But the COVID-19 pandemic actually showed us that vaccines can be developed much faster.” Researchers are looking at how to accelerate the development of life-saving treatments with AI.
Building trust
The use of AI in healthcare needs to be guided by a set of core principles, says Dr Priyadarshini. This is to ensure that AI not only augments human expertise but does so in a way that is ethical, secure, and trustworthy.
“It’s important for AI to be fair”, so that no segment of the population will be excluded from its benefits. AI in healthcare also must respect privacy and security, ensuring that patient data is protected at all times. Transparency is another vital aspect, Dr Priyadarshini stresses. “AI shouldn’t be a black box. Both doctors and patients should be able to understand how the model is generating insights.”
These principles are critical in building trust and accountability. “If an AI model fails, it’s not because AI itself is flawed. The issue often lies in the data it was trained on, which may be incomplete or insufficient,” she explains. This is where the importance of constant evaluation comes into play. “You need to ask, ‘Is it working? If not, do we scrap it or improve it?’”
Longer and better lives
Dr Priyadarshini shares an example of how AI can enhance patient care beyond the ICU. “In the ICU, patients receive round-the-clock, highly attentive care, but once they move to non-ICU settings, the patient-to-nurse ratio increases, sometimes one nurse for every 10 patients.” This is where vital signs can start to deteriorate unnoticed, and doctors may not always be available to intervene immediately.
To address this challenge, Microsoft and its partners implemented an innovative alarm system in Singapore hospitals. “When a patient’s vitals begin to dip, the system automatically triggers alarms that notify doctors and nurses on their phones, alerting them that the patient needs immediate attention,” Dr Priyadarshini says.
This proactive AI-powered solution has already been rolled out as a pilot across 11 hospitals, with impressive results. It was able to reduce the number of heart attacks – or what is known as code blue situations – because the staff could intervene earlier.
“Many patients don’t die from the disease they were initially admitted for,” says Dr Priyadarshini, “as complications like kidney or heart failure can arise when care is delayed.” By catching these critical issues early, AI is helping to prevent deterioration, improving patient outcomes and saving lives. She adds that this simple AI model uses data already being collected in hospital wards, making it a scalable and accessible solution that significantly enhances patient care.
AI’s potential goes beyond just managing critical care. As people live longer, maintaining a good quality of life becomes crucial. In Singapore, the average lifespan is up to 85 years, but the quality of life tends to decline after the age of 75, Dr Priyadarshini notes.
AI can then play a proactive role in helping people live healthier lives by nudging them toward good habits and healthier choices. From monitoring lifestyle factors like diet and exercise to providing personalised health recommendations, AI can encourage individuals to adopt behaviors that prevent illness in the first place. This helps to “extend not just life, but also the enjoyment of life itself,” Dr Priyadarshini says.
For more information on NUS-ISS Executive Education Programme in Digital Health, visit here.