Around this time last year, the 2021 AI in Healthcare Survey was released. The results showed growth in natural language processing (NLP), clinicians becoming primary users of AI technology, and a preference for companies using their own data to validate models, among other findings. And while the results didn’t change significantly from last year, there are several interesting trends that have emerged over the last 12 months.
The 2022 AI in Healthcare Survey results are in, and they shine light on the trends, challenges, and use cases driving health care AI forward. Factors like the global pandemic, aging populations, and rising health care costs have all contributed to how the industry is being shaped and where AI technology and tools can provide the most value. So, here’s what more than 300 respondents from over 40 countries are experiencing in their AI programs – and three key findings that stood out among the rest.
Data Annotation Becomes a Foundational AI Technology
When asked what technologies they plan to have in place by the end of 2022, technical leaders cited data integration (46%), BI (44%), NLP (43%), and new this year, data annotation (38% ). Text is now the most likely data type used in AI applications, and the prioritization of data annotation indicates an uptick in more sophisticated NLP technologies.
With this comes more sophisticated NLP use cases, such as clinical decision support and medical policy assessment. The pandemic has accelerated many areas of medical research, and drug discovery is no exception. This will become an even more critical area as we continue to adapt to new COVID-19 variants and consider how to prepare for the next pandemic. Better use of AI technology will be essential in helping with this challenge.
AI Users Shift from Data Scientists to Domain Experts
When asked about intended users for AI tools and technologies, over half of respondents identified clinicians (61%) as target users, and close to half indicated that health care providers (45%) are among their target users. Additionally, a higher rate of technical leaders cited health care payers and drug development professionals as potential users of AI applications.
As low- and no-code solutions proliferate, it’s likely that the shift from data scientist to domain expertise will continue in health care and beyond. This is an important part of the AI evolution. When doctors and specialists who have clinical expertise are paired with AI competency, the potential for the technology’s value grows exponentially. Democratizing AI and machine learning will open the doors for more use cases and more benefactors.
Cloud Providers Are Overshadowed by Open-Source Software
When asked what types of software respondents are using to build their AI applications, the most popular selections were locally installed commercial software (37%), and open-source software (35%). With the emphasis on open-source software came a 12% decline in use of cloud services (30%) from last year’s survey results (42%).
However, to the extent that AI increasingly relies on open-source technologies and public cloud providers, escalating security concerns are likely to have a substantial impact on both. Health care infrastructure has become one of the top targets for cyber criminals. In fact, over half of internet-connected devices used in hospitals have a vulnerability that could put patient safety, confidential data, or the usability of a device at risk, according to new research from Cynerio.
As security concerns indicate, it’s important to keep the shortcomings of AI in mind as development continues. However, for health care and many other industries, the rewards far outweigh the risks. That said, it’s encouraging to see that use cases are becoming more sophisticated, users are broadening beyond technical titles, and open-source solutions are more prevalent. These are all good signs for AI maturity, and similar to years past, there will be even more progress as we approach 2023.