Artificial intelligence (AI) is poised to change how many organizations do business. Hospitals are no exception to the rule. Advances in technology will help facilities provide better care in many ways, and this will likely happen sooner than later.
In fact, some hospitals are already taking advantage of the benefits of AI right now. Here are two real-life applications of AI technology that are currently impacting care delivery.
Actionable feedback
Patient satisfaction scores are becoming more critical for reimbursement. With that in mind, it’s important to find out patients’ impressions of their hospital experience so any issues that could impact scores can be corrected.
More patients are reviewing their hospital stays on sites like Yelp. But it can be difficult for hospitals to find, compile and analyze all the different reviews about them on various sites. One new AI platform aims to change that.
As written in an article from Forbes, NarrativeDX was created in 2014 by two developers. The cloud-based system helps hospitals see what people are saying about them in real time. Its goal is to help hospitals get actionable insights from patients’ online feedback by scanning review sites, social media and other platforms where patients discuss their experiences.
NarrativeDX uses a natural language processing system to pull out words and phrases dealing with patients’ concerns. Using these words, it makes recommendations for how facilities can solve patients’ problems and address issues.
Using data analytics, the AI tool can highlight different trends in patient feedback across multiple channels, making observations that humans might not notice.
It’s already helped several hospitals so far. Facilities that implemented improvements with help from NarrativeDX have generated an additional $50 million in revenue from patient referrals, and some have even increased their bottom line by 50%.
Hand-washing observation
AI’s helping other hospitals solve the age-old issue of hand hygiene.
Although it’s one of the best defenses against spreading bacteria, clinical staff aren’t always remembering to wash their hands when caring for patients. Studies have shown that adherence to hand hygiene protocol increases when clinicians are monitored.
Since hospitals can’t always have a person on hand to watch doctors and nurses, researchers from Stanford University have come up with the next best thing: cameras with AI vision. Per an article in Engadget, these smart cameras were used to observe hospital wards and highlight behaviors that could spread infections.
During a pilot project, cameras were placed in patient rooms and hallways, and near hand sanitizer dispensers throughout two wards of a hospital. The cameras took video over an hour during a busy period. Footage was both analyzed in real-time and after the fact by the camera’s AI vision algorithms.
The accuracy of the algorithms was tested against human eyes. Humans were more accurate in figuring out whether clinicians washed their hands when reviewing video played after the fact. However, the AI cameras were more accurate in real time compared with people who were secretly observing hand hygiene habits as the cameras recorded.
To test whether these results hold up in different environments, researchers will be setting up these smart cameras in three different hospitals for a yearlong period. The goal is to see whether AI will make a lasting difference with both improving hand hygiene and lowering infection rates.
If so, the cameras may become more widespread, as they’re a relatively cost-effective alternative to having human staff members monitor every area of a hospital.
Looking ahead
The capabilities of AI may intimidate some people, but forward-thinking hospital executives would be wise to follow developments related to AI and health care. It’s also smart to talk with IT to find out if your hospital will be able to adopt AI seamlessly down the road.
Embracing new technology and effectively incorporating it into clinical workflows is essential to thriving in an environment focused on quality and patient outcomes.
