Omnia Health is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Collaboration is key to the widespread adoption of AI for healthcare

Article-Collaboration is key to the widespread adoption of AI for healthcare

AI vendors and equipment manufacturers need to work with each other and healthcare providers to design products that meet clinicians’ needs

Artificial intelligence (AI) vendors are partnering with healthcare providers to identify practical challenges and consider AI technology as a potential solution. This type of collaboration is crucial to successful AI adoption because the technology is designed to tackle real-world healthcare problems. A myriad of examples was displayed at the 2020 Radiological Society of North America (RSNA) annual meeting.

The next 5 to 10 years will be pivotal for the expansion of the healthcare AI software market. Omdia forecasts that the healthcare AI software market will grow from about $800 million in 2019 to nearly $11 billion in 2025, representing a 54% CAGR.


Figure 1: Global AI software revenue for healthcare; Source: Omdia

In recent years, healthcare practices have dramatically increased AI spending. The latest annual (2020) Information and Communications Technology Enterprise Insights in the Healthcare Industry (ICTEI) survey found that this trend will likely continue. Nearly 60% of respondents reported plans to increase AI spending in the next 18 months. The Omdia Artificial Intelligence for Ultrasound Survey (AIUS) 2020 and ICTEI 2020 survey identified many factors driving the increase in AI spending, including improved diagnostics, time savings, and government initiatives.  The unforeseen health crisis caused by the coronavirus disease (COVID-19) may be the greatest market accelerator in the near-term.

Despite several tailwinds, currently, healthcare AI is used primarily in advanced hospitals, universities, and research facilities in mature markets, like the US and Western Europe. Meanwhile, much of the medical world has limited exposure to AI, with lingering doubts that technology is a good use of budget. It is incumbent upon AI vendors and equipment manufacturers working with each other and healthcare providers to design AI products that realistically meet clinicians’ needs and share some of the financial and logistical burden associated with purchasing and implementing AI.

A confluence of factors is promoting rapid AI adoption

The Omdia AIUS 2020 explored the internal factors driving the adoption of AI. Over 80% and 60% of respondents identified “improved diagnostics” and “time savings” as top drivers for AI adoption, respectively. These results are consistent with the survey’s findings related to the current impact of AI and industry expectation of AI’s ability to improve the accuracy and efficiency of medical imaging. Vendors at RSNA 2020 were promoting the diagnostic and time-saving capabilities of AI products. For example, CureMetrix showcased the results of a study that found that its AI software improved radiologists’ diagnostic accuracy and specificity by 25% and 34% respectively, and reduced the time required to read normal exams when interpreting mammograms (Source: CureMetrix).

Omdia’s ICTEI 2020 survey identified that external factors, including “federal healthcare industry-wide initiatives”, are crucial in driving AI adoption. In 2019, the UK government announced a £250 million budget to establish a new national AI laboratory dedicated to developing solutions to address the nation’s severely strained healthcare system. Similar initiatives have been implemented in other Western European countries, as well as the US, China, Japan, and South Korea.

One of the most significant external drivers of AI adoption in the healthcare industry in the short- and long-term will be the COVID-19 pandemic. Efficient and effective COVID-19 diagnoses is of upmost importance during the pandemic, which has accelerated the development of AI software in the healthcare sector.  Further government investment in AI-based drug and vaccine research, medical imaging, and machine learning tools for patient screening, triage, and monitoring is also expected. While the impact of the pandemic will be temporary, COVID-19 will serve as a litmus test for AI technology. At RSNA 2020, RADLogics showcased its latest iteration of AI solutions designed to detect symptoms of COVID-19 from CT and X-ray images (Source: RADLogics).


Figure 2: COVID-19 forecast effect on AI software industries; Source: Omdia

Stakeholders must collaborate to overcome barriers

To reach its full potential in the healthcare market, AI vendors must overcome daunting barriers to adoption. The AIUS 2020 and ICTEI 2020 survey identified that information technology (IT) and workflow integration and education and training were top barriers from the healthcare provider perspective. To overcome these challenges, healthcare equipment and AI vendors should increase ongoing support with clients’ IT departments, to ensure that AI technology can be integrated effectively. AI vendors should also improve and increase engagement with healthcare providers to support education and training, which has shown to significantly increase the likelihood of future AI adoption.

Cost is also a major barrier to healthcare AI adoption. Over 50% of AIUS 2020 survey respondents identified that the cost of AI was too high. AI solutions may need to be simplified, eliminating less important features, to bring down the cost to meet customer expectations. Conversely, with proper education and training, vendors may be able to maintain or increase prices by justifying the value of AI. During the early stages of AI adoption, it might be necessary for vendors to share some of the capital burden associated with adoption by utilizing nontraditional business models. The 2020 AIUS and ICTEI survey results identified a discrepancy between the business models currently used for AI and end-user preference. More than 90% of ICTEI respondents identified a preference for an operating expenses (OPEX) type AI business model; however, the AIUS results showed that over 60% of respondents adopted AI through one-time purchases. Subscription-based business models reduce healthcare providers’ financial risk and provide new opportunities for AI software to demonstrate value and justify the long-term investment.

Collaboration between AI vendors, healthcare equipment manufacturers and healthcare providers is essential to further AI development and adoption. At RSNA, NVIDIA announced the launch of its Inception Alliance program which will facilitate partnerships between medical imaging AI startups, GE Healthcare and Nuance (Source: NVIDIA). The program will help to fast-track the development of AI for healthcare and link software development with the medical imaging community. Collaborations like Inception Alliance will help healthcare AI take the next step in its expansion as advanced technological developments will be guided by real-world healthcare challenges.

Read part one of the article here

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.