1. Personalized Preventive Care
The traditional approach to preventive care uses broad, universal categories like age and sex to determine what medical screenings and preventive measures people should receive. While this approach works on a large—and mostly probabilistic—scale, it overlooks individual risk factors, such as genetic profiles and specific biomarkers. Humans can’t possibly retain and recall all the data that affects each individual patient, but they will be able to apply the totality of information using GenAI. These tools can almost instantaneously analyze vast amounts of data to identify those individuals at highest risk for serious conditions like cancer and heart disease and recommend targeted approaches to improve patient outcomes.
2. Reducing Diagnostic Errors
Traditionally, doctors rely on memory and experience to diagnose new symptoms, drawing from their medical training and past cases. However, this approach risks overlooking rare or complex conditions, and it is vulnerable to cognitive errors like confirmation and proximity bias. GenAI offers a powerful solution by integrating vast amounts of medical data—combining patient history, symptoms, and real-time imaging—and correlating it with comprehensive medical literature, including obscure case reports. In the future, combining a dedicated clinician with a generative AI application will produce more accurate diagnoses than either alone.
3. Enhancing Doctor-Patient Interactions
In today’s healthcare, doctors often spend more time inputting data into electronic health records (EHRs) than engaging with patients, leading to impersonal and transactional experiences. GenAI is changing this dynamic by automatically transcribing and organizing doctor-patient conversations into accurate, high-quality EHR entries. This technology not only frees up to two hours a day for clinicians but also improves the quality of care and helps reduce burnout.
A Fourth Opportunity: Accelerating Medical Research
In addition to Dr. Topol’s three points, I’d add a fourth: the ability of GenAI to accelerate research. In medical science today, it can take years to gather enough data to drive meaningful advances. GenAI can dramatically shorten this timeline by analyzing vast amounts of patient data quickly, leading to faster breakthroughs and more timely application of new treatments.
Clinical research conventionally starts with a question, followed by lengthy data collection and analysis. This approach is time-consuming and limited by the volume of data that researchers can analyze and manage. GenAI alters the calculus by enabling doctors to sift through enormous datasets. Today, U.S. hospitals produce up to 50 petabytes of data each year, 97% of which currently goes unused. By mining this data, GenAI will be able to uncover patterns and insights that would take years to find with traditional methods. One of the first practical applications will be identifying hospitalized patients who are likely to deteriorate over the next 24 hours, allowing clinicians to intervene earlier and potentially save lives.
Challenges and Risks
Of course, breaking the rules of medicine comes with challenges. Security is a major concern, especially when clinicians use generative AI for EHR data entry. However, the reality is that this danger already exists in the current EHRs, which can easily be hacked, but fail (alone) to offer the advantages that GenAI solutions will provide.
The evolution of medical technology always includes trade-offs. The advent of CT scans, MRIs and laparoscopic tools, for example, led doctors to lose their skill in physical exams, but the lives saved by these innovations are undeniable. No clinician would go back to the past.
Within the next five years, Dr. Topol predicts that GenAI will become a standard tool for creating electronic health records (EHRs). Other applications will follow soon after. I’m confident that the old rule “the doctor knows best” will be replaced by a new reality—one in which the best outcomes come from a collaboration between a dedicated clinician, an empowered patient and GenAI. Together, they will achieve more than any of the three could accomplish alone.
Neither Providers nor Patients will be able to avoid Chatbots and other AI assistants
Dr. Robert Pearl is the author of “ChatGPT, MD: How AI-Empowered Patients & Doctors Can Take Back Control of American Medicine.” All profits from the book go to Doctors Without Borders. Fixing Healthcare is a co-production of Dr. Robert Pearl and Jeremy Corr. Subscribe to the show via Apple, Spotify or wherever you find podcasts. Join the conversation or suggest a guest by following the show on Twitter and LinkedIn.
No comments:
Post a Comment