Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Five pitfalls of implementing an AI initiative in healthcare

As more hospitals turn to artificial intelligence to improve patient care and daily operations, missteps like overlooking security and compliance can cause an AI project to fail.

Artificial intelligence has certainly taken center stage in the last 12 months, especially with ample examples...

of its success in different industries outside of healthcare. However, some hospitals and health plans have highlighted their own successes with AI initiatives around patient care and satisfaction improvements, as well as improved staff efficiencies. Given the number of available options to choose from that range from AI as customer service bots to detecting abnormalities in medical images, there is plenty of AI to go around.

Instead of hospitals embarking on the latest or hottest AI trend that is advertised in front of them, some in health IT recommend hospitals consider the lessons learned from early adopters to ensure their AI initiative is successful. Here is a list of common pitfalls that hospitals should keep in mind as they begin their next AI initiative and journey.

Embarking on too many AI projects at once

AI comes in so many assorted flavors that can benefit different departments within a hospital. Whether a hospital is looking to use AI for medical imaging classification, scheduling or a virtual assistant in patient rooms, it must be aware that all of these require numerous resources. To increase the success rate of AI projects, IT and other stakeholders must focus on a small set of AI-based initiatives and see them through to ensure they are fully utilized and have met their objectives. 

Focusing on the technology, not the outcome

While the technology used in AI can be intriguing to many in the technical field, the focus should remain on delivering the specific outcomes. IT and the members involved in the AI project must avoid getting distracted by the software tools and clearly define the success measures needed for the project. For example, if improving patient satisfaction is the goal, measuring success might include surveying patients before and after an appointment. 

Making the AI initiative just another IT project

IT typically leads the selection and implementation processes for technology initiatives. Projects related to infrastructure and other data center components may not require the participation of other departments within a hospital. However, when it comes to deciding on an AI platform, IT should not be the sole decision-maker. Input from other departments that will interact with the technology can help determine if the AI tool addresses their needs or those of patients.

Limiting the data AI can access

With the exception of a few AI-based tools like digital voice assistants or autonomous robots that deliver medications and clinical supplies within a hospital, most AI tools consume tremendous amounts of data to help it learn and improve the accuracy of decision-making. Some of the AI platforms available for clinical analysis require direct access to patient's charts and are able to analyze millions of patient's charts that may include labs, images, medical history and other relevant data.

Deploying AI technology without adequate compliance and security review

There are several concerns around privacy and security when it comes to several of the AI services in the marketplace. Depending on the AI technology itself and whether data is being transmitted out to a centralized processing server outside the hospital, security and privacy can't be overlooked. Tools like digital voice assistants can add value if they are rolled out within a hospital setting, but because of the way they capture voice data and process it outside of the hospital environment, as well as a general lack of HIPAA compliance, these solutions simply can't and won't be compliant in their current configuration.

Similar to any other project a hospital is engaging in, an AI initiative will require the collaboration of several departments to ensure it is successfully implemented. As more hospitals implement AI projects, more lessons learned can further be shared among IT departments and other specialties. But for now, many are still discovering what AI can do for them.

 

This was last published in March 2018

Dig Deeper on Health care business intelligence systems

Join the conversation

1 comment

Send me notifications when other members comment.

By submitting you agree to receive email from TechTarget and its partners. If you reside outside of the United States, you consent to having your personal data transferred to and processed in the United States. Privacy

Please create a username to comment.

What do you think is the biggest pitfall to implementing an AI initiative in healthcare?
Cancel

-ADS BY GOOGLE

SearchCompliance

SearchCIO

SearchCloudComputing

SearchMobileComputing

SearchSecurity

SearchStorage

Close