Healthcare AI Trends: Managing health inequities and enterprise growth is the focus for 2022 and beyond

health inequities

Healthcare AI Trends: Managing health inequities and enterprise growth is the focus for 2022 and beyond

Healthcare has fundamentally changed. Looking back on the last two years, COVID has pushed our industry in ways we could never have imagined. We’ve seen…

  • Redesigned timelines for vaccine development
  • Accelerated use of telemedicine
  • The growth of technology solutions to improve care
  • Bias, racial disparities, and inequity in care
  • A backlog in preventative testing and follow up care

These changes have been both good and bad for patient outcomes. Our challenge now is to take a hard look at the healthcare industry and find ways to fix what’s broken while ensuring the positive developments continue. These past two years have been a catalyst for redefining healthcare priorities in the technology space.

An increased focus on using artificial intelligence (AI) in healthcare has emerged as a growing technology priority. Health systems are looking to AI to solve both old and new problems. This got me thinking about the year ahead and what we might see in the healthcare AI space. A quick Google search highlighted a broad scope for the use of AI.

The current state of AI in healthcare – 6 important trends for 2022 provides an excellent overview of the different ways AI can play a valuable role in healthcare.

Regulatory changes, technology advancements, and consumer needs are driving forces in the growth of AI. Beyond COVID: Biotech and Health Care Trends to Watch In 2022 looks at artificial intelligence, diagnostics, and medical fintech as game-changers in the coming year.

Looking closer to home, I asked members of the Ferrum Health team their thoughts on the use of healthcare AI in the coming year.  Here’s what I learned.

Pelu Tran, Ferrum Health Co-Founder and CEO

There is a dawning awareness across healthcare that technology can be both the cause and the cure for health inequity. As a result, providers, patients, and regulators are all demanding more accountability and visibility into the performance and bias of the AI tools they are using, ultimately driving greater trust and more rapid adoption of these tools by clinicians and patients.

Looking forward, I think the adoption of single-point AI tools will not be sustainable. The time, cost, and stress on a hospital’s clinical and technical teams will be overwhelming as their use of AI grows. Therefore, the trend will be for health systems to first invest in the validation and monitoring infrastructure needed to support their ecosystem of clinical AI tools. This investment will ensure their long-term success in using AI.

COVID has also made us acutely aware of the disparities in healthcare. My hope is this realization, and the work being done to fix these inequities is a learning that drives actual change in our industry. As part of the PandemicX Accelerator, we’ll be working with the U.S. Department of Health and Human Services to address health inequities and mitigate the effects of the COVID-19 pandemic. I believe AI will play a significant role in addressing these challenges.

Providers, patients, and regulators are all demanding more accountability and visibility into the performance and bias of the AI tools they are using, ultimately driving greater trust and more rapid adoption of these tools by clinicians and patients.

Ken Ko, Ferrum Health Co-Founder and CTO

Looking back at the past ten years, it is no surprise to see the cloud has been top of mind for everyone. There have been major shifts within companies and across multiple industries that make fundamental changes to their software architecture being delivered and designed to take advantage of this once-new environment. What started off simply as “other people’s computers” has demonstrated time and time again that elasticity and on-demand scaling is an enormous benefit for enterprises, allowing their teams to focus on what they’re delivering rather than the how.

A modern trend gaining traction is the shift away from this centralized cloud dependency. This movement to fan out and perform more computation at the edge of a network results in computation and data residency to be much closer, and sometimes adjacent to, the data sources. This can be seen with the recent offerings and service offerings from the major cloud, platform, and infrastructure providers.

Concurrently, in the world of AI and machine learning, developers and data owners are becoming more familiar with the notion of organizational data lakes. These are centralized repositories that contain structured and unstructured data for access and sharing. The organizational and operational risk involved if a leak were to happen has rightfully created an air of paranoia.

What we are going to see in this next year is the nexus of these two ideas. Bringing compute closer to the edge within companies’ data centers allows IT teams to administer these nodes with the same security policies and auditing used today. The end result is greater data privacy and data residency policies that are favorable to individual organizations, which is especially vital for healthcare systems. Creating this environment enables healthcare systems to rapidly deploy tools to empower their staff of AI-powered humans in all the service lines.

healthcare AI

Elie Balesh, MD, Radiologist and Ferrum Health Medical Director

AI Winter is coming, but Spring isn’t too far away.

Healthcare is notoriously slow at implementing new digital technologies, lagging other industries by a decade or more. What has further inhibited the adoption of AI in healthcare, though, is the lack of product-market fit: developers solving problems that physicians do not perceive exist and charging premium prices for them in a budget-conscious, post-pandemic healthcare economy.

I anticipate the AI market will consolidate and recalibrate, shifting from the lofty goal of machine-generated clinical diagnosis to the more mundane but valuable use of AI to optimize operations. In a radiology department, examples may include tools to automate examination selection, scheduling, protocoling, acquisition, artifact reduction, post-processing, and communicating results and guideline-based recommendations for follow-up. AI can also offer excellent value post-diagnosis in quality assurance and opportunistic population health screening programs. These types of workflow solutions will resonate with most stakeholders in radiology much better than CAD type products that claim to out-perform radiologists in some manner.

With respect to finances, AI startups are either resource-strapped or over-funded and beholden to investors to grow fast and generate revenue to justify high valuations. In either case, firms have three go-to-market strategies: sell directly to hospital systems, partner with a distribution platform, or be acquired by a large OEM and integrated as a feature of a scanner or PACS. Long sales cycles and bureaucratic red tape will cause direct sales to wane as firms make better use of their resources for product R&D. In 2022 we will see a growing number of platform partnerships, with companies such as Ferrum providing a solution to the “last mile” problem of integrating with enterprise health systems.

AI can also offer excellent value post-diagnosis in quality assurance and opportunistic population health screening programs.

Rob Brill, Ferrum Health Head of Client Development

I have the privilege of working directly with our customers, and I’m seeing a shift toward the enterprise use of AI. Single point AI solutions that have historically been the starting point for a health system to use AI are becoming challenging to manage. The work needed to execute, validate, and monitor each individual AI algorithm is overwhelming the health system’s resources.

The trend is moving toward a tool or solution that allows health systems to easily manage a variety of AI algorithms used together to address specific therapeutic areas such as lung, breast, or bone health.

What are your thoughts on the use of healthcare AI? What trends do you see emerging?

Drop us a note; we’d love to hear your thoughts on the future of healthcare AI.

 

Kathleen  Poulos

Kathleen Poulos

Kathleen is a registered nurse with a digital marketing background, a love for using technology to solve healthcare challenges and a passion for improving patient care.

Contact Us

CASE STUDY

ARA Health Specialists

Use the button below to download your free case study and learn how our approach to validation has improved the number of clinically significant findings in AI software.

CASE STUDY

Sutter Health

Use the button below to download your free case study and learn how our approach to validation has improved the number of clinically significant findings in AI software.