AI in Healthcare – Part 1

After I finished my doctorate in health administration (at age 69), I wrote Health Attitude: Unraveling and Solving the Complexities of Healthcare, a book on what I had learned about the American healthcare system. That was 2015. Since then, there have been medical advances, but in my opinion, the healthcare system is further behind. Healthcare insurance and pharmaceutical companies are seeing record profits, hospitals are struggling to breakeven, and patients remain confused about how to navigate the system to manage the health of themselves and their families. When it comes to automation of the administrative aspects of healthcare, the industry is way behind. I see AI as a breakthrough which can enable healthcare to catch up.

The use of AI in healthcare has roots that go back further than you might think. The groundwork for AI in medicine began in the 1950s alongside the development of the field of AI itself. By the 1970s, we saw the emergence of early AI applications in healthcare, like MYCIN, an expert system developed at Stanford University which assisted doctors in diagnosing and recommending treatments for bacterial infections. However, MYCIN had limitations. It required extensive data entry by the physician, lacked the ability to learn and adapt over time, and couldn’t handle the complexities of real-world clinical cases that often involve multiple factors beyond infections.

There was limited progress in the 1980s and 1990s. While research continued, the 80s and 90s ran into significant limitations in processing power and data availability. This hindered the widespread adoption of AI in healthcare practices. In essence, AI was stuck in the mud. In the following paragraphs, I will explain the key factors and how they facilitated the boom in AI.

Things changed dramatically in the 2000s and especially in the last couple of years. There was a handful of factors which changed everything for AI in an exponential way. First was the introduction of awesome computer power. Santa Clara, CA based Nvidia started with gaming graphics cards (GPUs). In 2006, they opened up the power of these GPUs for general computing, including AI.  These GPUs were perfect for the massive data processing needed in AI, especially deep learning. By making this powerful hardware accessible, Nvidia supercharged AI research and development, leading to the explosion of AI innovation we see today.

Equally important was to have a reliable and fast Internet. Over the past decade, Internet speeds have skyrocketed from millions of bits per second (Mbps) to billions (Gbps) thanks to fiber optic expansion.  This, along with improved infrastructure and provider focus on uptime, has led to more reliable connections. AI’s rise is deeply tied to this faster, more reliable Internet. The massive datasets and complex calculations required for AI training and operation need this boost.  Think of it as a powerful car needing a high-speed highway. Increased Internet speed allows AI to process information much faster. Additionally, a reliable Internet reduces disruptions during training or operation, ensuring smoother AI development and real-world performance.

AI thrives on data, and the vast reserves of information needed for training complex algorithms used to be a major bottleneck. However, the emergence of cheap and nearly infinite cloud storage has revolutionized the field. The readily available storage allows researchers to accumulate and access enormous datasets, fueling the development of ever more sophisticated AI models.  With data no longer a limiting factor, AI can now process and learn from massive amounts of information, leading to significant advancements in areas like image recognition, natural language processing, and scientific discovery.

The allure of AI has sparked a student migration toward the epicenters of this revolutionary field. Top universities with robust AI research programs and strong ties to industry are witnessing an influx of talented students. These institutions offer not just cutting-edge theoretical knowledge but also practical experience through collaborations with leading AI companies. This proximity to the forefront of AI development fuels student enthusiasm, providing them with opportunities to work on real-world projects and learn from the brightest minds in the field. The chance to be at the heart of this transformative technology is proving to be a powerful magnet for ambitious students, shaping the future landscape of AI with their skills and passion.

The recent explosion of AI can be attributed to this perfect storm of advancements. Faster computers with more processing power provide the muscle for complex AI algorithms. High-speed Internet acts as the nervous system, allowing for the rapid exchange and sharing of massive datasets. Meanwhile, nearly infinite and cheap cloud storage eliminated data limitations, allowing researchers to hoard and access the information needed to train ever-more sophisticated AI models. Finally, the wave of brilliant computer science researchers flocking to the field fueled the immense potential of AI. This confluence of factors – powerful hardware, high-speed data transfer, vast data storage, and top minds – has facilitated the development of faster and more efficient algorithms, the true engines that make AI tick. These advancements allow AI to process information at unprecedented speeds and uncover hidden patterns within the data, leading to breakthroughs in various fields, including healthcare.  

The AI sector is experiencing a boom in investment. While venture capital funding dipped slightly in 2023 compared to 2022, it still reached impressive heights. According to Crunchbase data, generative AI and related startups attracted nearly $50 billion in funding rounds. This indicates a strong investor confidence in the potential of AI, with established players like OpenAI and Anthropic securing large sums alongside numerous emerging companies. This influx of capital fuels innovation and fuels the growth of the AI landscape, with the exact number of startups constantly evolving due to the fast-paced nature of the field. While the exact number can fluctuate due to the dynamic nature of the startup world, estimates suggest there are currently around 67,000 to 70,000 AI startups worldwide. This data comes from various sources like Tracxn and Exploding Topics, highlighting the vast and ever-growing landscape of AI innovation.       

AI’s market potential is booming and is expected to reach $trillions by 2030 with a growth rate exceeding 28%. This surge is driven by AI’s adoption across all industries, fueled by advancements in the factors I discussed earlier. Generative AI and automation are fueling healthcare advancements.

In “AI in Healthcare – Part 2”, I will discuss the specific advances in healthcare. I will highlight the positives, but also the risks. I will offer examples in all aspects of healthcare.

Note: I use Gemini AI and other AI chatbots as my research assistants. AI can boost productivity for anyone who creates content. Sometimes I get incorrect data from AI, and when something looks suspicious, I dig deeper. Sometimes the data varies by sources where AI finds it. I take responsibility for my posts and if anyone spots an error, I will appreciate knowing it, and will correct it.