When people think of AI in healthcare, they might imagine a dystopian future where human physicians are replaced by robots, removing any semblance of humanity and emotion in a field where bedside manner is essential. However, the reality of the role of artificial intelligence in healthcare today is far less grim. Amongst the slew of proven benefits in the clinical domain, healthcare AI has cemented itself as a sidekick to physicians. For Star Wars fans, think about Luke Skywalker and R2D2. While Luke was the hero of the story, it was R2D2 who opened doors that Luke otherwise would have to find an arduous way around. Think of AI in the same regard – instead of added “boots” on the ground, it acts as extra “circuits” on the ground for clinicians and health systems, swiftly navigating around obstacles that could have otherwise been dead ends in clinical pathways, simplifying the day-to-day life of a physician.
In this blog post, we’re going to go over the history of AI in healthcare, the many roles that AI can play in today’s healthcare system, the benefits of AI in healthcare and more.
The History of AI in Healthcare
With some of its first definitions being laid out in the 1950s, there was no way of telling just how far AI would come. It has evolved from a far out, unrealistic idea to an integral part of modern living. Whether that be algorithms feeding us suggestions on Netflix and YouTube, making recommendations based on inputs from smartwatches or even impacting clinical workflows, there was no way of knowing just how relevant AI would become to all of us.
Several variables contributed to the expansion of AI powered healthcare systems:
- Notable strides in computing power, allowing for vast sets of data to be processed more quickly
- The widespread implementation of EHRs
- Improvements in natural language processing (NLP) and computer vision
The Role of AI in Healthcare Today
With the advent of popular generative AI models like ChatGPT taking the world by storm, the expectations of what AI can and ought to be able to do have skyrocketed. In some ways, this is warranted. In others, expectations should be tempered.
AI is Meant to Augment Physicians, Not Replace
Simply put: AI cannot and should not replace a physician. Experts have predicted that AI would phase out the need for radiologists and, while AI can drive automation, not all industries where AI can be applied are equal in that sense. Manufacturing, for example, is not like healthcare and doesn’t operate with the same stakes at hand.
What AI Can Achieve in This Role
To say that the limit of AI’s role as an augmentor and not the end-all solution to replace physicians is not to minimize the limitless impact it can effectuate on healthcare today. Take Envision Healthcare, for example. In studying the impact of AI on identifying lung blood clots, they observed that while AI might have been more effective at identifying true positive cases, the radiologist was stronger at identifying true negatives, concluding:
“When the exceptional AI sensitivity is coupled with the superior radiologist specificity, both parameters are optimized resulting in substantial gains in accuracy which are clearly reflected in the data.”
In this complementary role, AI can achieve numerous objectives in clinical workflows:
The Healthcare AI Explosion
Understanding the role AI has, can and will continue to play in healthcare is underpinned by its rapid adoption and its explosive growth vis a vis AI/ML enabled devices that have received FDA clearances. On May 13, 2024, the FDA updated its number of such devices to 882. This latest revision alone provided an additional 191.
As stated by Aidoc CEO Elad Walach, “Our AI is currently analyzing over 2 million patients a year. With other strong companies[…]I believe that up to 30% of patient care is already benefiting from the utilization of clinical AI solutions.”
The Role of AI in the Future of Healthcare
While the first wave of AI adoption and implementation hinged on single use-case applications, the role of artificial intelligence in healthcare has changed. Health systems that desire a long-term, sustained and scalable AI strategy must approach AI from a platform perspective. That’s to say that a multitude of algorithms are being fed through a protocol-agnostic, single user interface, avoiding a host of workflow and management challenges that come with 30-40 algorithms operating in isolation. That would explain why, when looking at today’s healthcare AI conversations, we’re seeing vendor consolidation as a predominant topic. Smaller, use-case focused developers are going to look to leverage platforms that can handle data normalization, workflow integration all while acting as a single point of contact for health systems.
For a deep-dive AI conversation, you can listen to Elad Walach, CEO at Aidoc, discussing the founding vision for Aidoc and his vision for the future of AI in healthcare with the BackTable Podcast:
