The Future of AI in Healthcare: A Transformational Shift

With the rapid advancements in generative AI over the past two years, Artificial intelligence (AI) is no longer a futuristic concept in healthcare — it’s a rapidly evolving force that is set to reshape patient care, clinical workflows, and provider-patient interactions.?

To cut to the chase: One of my watershed moments was when I heard a Mayo Clinic physician with 35+ years of experience, whom I’ve known from my time at the Mayo Clinic, mention on stage at a Mayo Clinic Health System event last year that he is excited about how AI will be able to transform the physician patient interaction, allowing him to “practice medicine like I did in the 1990s — only better”.

Here’s my prayer: May this prolonged period of “technical interference” in the physician-patient relationship soon be a “what were we thinking” relic of the past and may we from now on use technology to truly enable the delivery of extraordinary care.

At the Telehealth T-Time event in December 2024, participants engaged in a thought-provoking discussion about the future of AI in telehealth and its potential to revolutionize healthcare delivery.?

The consensus? AI is poised to become pervasive, offering unprecedented opportunities while requiring careful implementation to maintain transparency and trust.

Here are the group’s conclusions and insights.

AI as a Game-Changer in Preventive Healthcare

One of the most compelling insights from the discussion was the role AI could play in preventive healthcare. Health insurers, especially Medicare, are at present “left holding the bag” as a result of poorly managed health due to lack of preventive measures that take hold.

Currently, healthcare systems struggle with maintaining consistent patient engagement, particularly in ensuring medication adherence, tracking health metrics, and addressing patient concerns before they escalate. The shortage of clinical staff exacerbates this challenge, limiting the ability to provide ongoing follow-up care.

AI-powered health assistants could bridge this gap by serving as digital accountability partners, ensuring patients remain engaged in their care plans. These AI tools can check in with patients, provide reminders, and surface critical concerns to human providers. The potential impact is immense: AI-driven follow-ups could reduce hospital readmissions, improve chronic disease management, and enhance overall patient outcomes.

A key observation from the discussion was that many patients hesitate to contact their doctors until their symptoms become severe. The introduction of AI assistants could alleviate this hesitation by providing an always-available, non-judgmental resource for health-related questions. These assistants could triage issues, offering basic guidance while escalating complex concerns to human clinicians, thus optimizing the efficiency of healthcare teams.

Personalization and AI-Driven Genomics

Beyond preventive care, AI holds significant promise in the realm of personalized medicine. Advances in AI-driven genomics could lead to highly individualized treatment plans tailored to a patient’s unique genetic profile. The ability to analyze vast datasets at an unprecedented scale means that AI can uncover patterns and predictive insights that would be impossible for human clinicians alone.

For instance, AI can help identify genetic predispositions to certain conditions, allowing for earlier interventions and more targeted therapies. This level of personalization has the potential to shift healthcare from a reactive to a proactive model, fundamentally changing how diseases are prevented and managed.

My favorite example of the wasteful nature of non-personalized care comes from Dr. Eric Topol’s book from 2014 “The Patient Will See You Now”: A drug that reduces mortality from 8% to 4% is 50% effective - but it does NOTHING for 96% of the people taking it (the 92% that would have never died and the 4% that still died). What if we can figure out which 4% of the population should have taken it (or even which 8% were at the highest risk).

My hope is that in the near future AI will take my genomic data and my medical history data and the latest research insights to develop for my "wellbeing quarterback" a personalized wellness plan that may include the right medications and the right supplements alongside a prescription of a diet and movement regimen optimized for my health.

Building Trust Through Transparency

Despite its vast potential, AI in healthcare must be implemented with transparency and accountability. A major point raised in the discussion was the necessity of avoiding a “black box” AI system—one in which decisions are made without clear explanations. Both providers and patients must be able to understand how AI reaches its conclusions, ensuring trust in the technology.

There is also a growing understanding that AI needs to be closely monitored and checked, e.g., as even algorithms seem to have a tendency to drift over time.

One suggested approach is integrating AI in a controlled, “sandbox” environment where its recommendations can be vetted before being fully deployed in clinical settings. This iterative process can help build confidence among healthcare providers who may initially be skeptical of AI-generated insights. Over time, as AI proves its reliability, its adoption will likely accelerate, leading to deeper integration into healthcare workflows.

While this will slow down the adoption of AI, it will still be vast improvement from the average of 17 years that it used to take for 50% of clinicians to adopt the proven insights from landmark controlled clinical trials into their practice (read that again: only 50% after 17 years. Mindboggling.).

The Accelerating Adoption of AI in Diagnostics

Perhaps one of the most striking revelations from the discussion was the rapid advancement of AI in diagnostic capabilities. A recent study cited in The New York Times found that GPT-4, a model over a year old at the time of the discussion, outperformed human physicians in certain diagnostic tasks. However, this study came with an important caveat: The AI was working with highly controlled data, carefully curated by trained professionals.

Even with this limitation, the findings suggest that AI-driven diagnostics are advancing at a much faster rate than previously anticipated. As models become more sophisticated and are trained on more diverse datasets, their accuracy and reliability will only improve. This shift could have profound implications for telehealth, enabling remote diagnosis and triage at an unprecedented scale.

On the other hand, I had a great conversation with a radiologist the other day, pointing to the cognitive (over)load if the AI handles all the easy cases and leaves the tricky ones to humans. It also greatly skews the experience-based bias for rare conditions to be, well, very rare.?

What Comes Next? Preparing for an AI-Integrated Future

The conversation at Telehealth T-Time underscored a crucial reality: AI in healthcare is not just coming — it’s already here, and it’s evolving quickly.?

Healthcare leaders must begin preparing for a future where AI is an integral part of clinical decision-making, patient engagement, and operational efficiency.

To ensure successful integration, healthcare organizations should:

Foster a culture of AI literacy among providers and staff to build confidence in AI-driven insights.?

Implement AI tools in controlled environments first, allowing clinicians to validate their accuracy and reliability.?

Prioritize transparency and explainability in AI systems to maintain trust with both providers and patients.?

Stay informed about emerging research and regulatory guidelines to navigate the ethical implications of AI adoption.

The future of AI in telehealth is brimming with potential, but its success hinges on thoughtful implementation, trust-building, and continuous adaptation. As the healthcare landscape evolves, embracing AI as a collaborative tool—rather than a replacement for human expertise—will be the key to unlocking its full potential.

What are your thoughts on AI’s role in telehealth? Share your insights and join the conversation as we navigate this transformative era together.

To receive articles like these in your Inbox every week, you can subscribe to Christian’s Telehealth Tuesday Newsletter.

Have you attended Telehealth T-Time: A Community for Telehealth Enthusiasts yet? This monthly, FREE event always features networking opportunities and three presentations by fellow telehealth experts. Check it out!

Christian Milaster and his team launch, expand, and grow Telehealth Programs for rural health centers, behavioral health agencies, health systems, schools, and libraries. Christian is the Founder and CEO of Ingenium Digital Health Advisors where his team and consortium of experts partner with healthcare leaders to enable the delivery of extraordinary care by accelerating the adoption of digital health innovation.

To explore how we can help your organization solve your challenges, contact Christian by phone or text at 657-464-3648, via email, or video chat.