Artificial intelligence tools are helping primary care doctors diagnose skin cancer, home care providers retain employees and hospital doctors get early warnings about potential patient mental health episodes in stories we're reading this week. 

Solventum announced it is now using AI to flip revenue cycle processes and address denials proactively. Through a new partnership with Ava, an enterprise AI platform, WellSky enabled a large home care provider to improve employee engagement and reduce turnover using tools integrated into electronic health records. 

San Diego State University researchers leveraged grant funding to outfit and test AI-driven mental health robots, while the use of an accelerated medical device in general practice highlighted the importance of post-market research.

Preventing denials within workflows

On Monday, Solventum and Sift announced a machine learning integration that intervenes in coding and pre-bill validation within the front end of the healthcare revenue cycle.

The new revenue integrity system is designed to help health systems reduce potential denials and ensure timely and accurate payer reimbursement, the companies said.

While healthcare organizations have cited claim denials as the greatest revenue cycle management challenge, moving to a proactive strategy can help them prevent denials in the clinical workflow, according to Garri Garrison, president of health information systems at Solventum, formerly 3M Healthcare.

"As healthcare systems continue to face increased write-offs due to the growing denials burden, the current reactionary approach of relying on claims data to reverse engineer denials isn't working," she said in a statement.

Community Medical Centers of Fresno told Healthcare IT News in March that it turned to AI to reduce claims denials. 

"By having this tool, we could stop claims on the front end, which gave us one last chance to fix any outstanding items that could trigger a denial," said Eric Eckhart, the provider's director of patient financial services.

Advancing mental health robots

With a $5 million grant from the Brown Foundation, researchers at SDSU are taking their robots, Pepper and Bernard, to the next level of emotional sensitivity. 

Dr. Aaron Elkins, the center's director, told NBC San Diego earlier this month that researchers are working on artificial intelligence that could act as an early warning system for a mental health event. Such personalized therapeutic robots could alert a doctor of the potential for a mental health episode.

"A machine might be able to see changes in your pupil, might be able to hear your voice pitch change or something happens that it senses that you're maybe in a certain state that needs a certain kind of intervention," Elkins said in the story.

The center has partnered with area hospitals and is testing the technology to diagnose diseases and disorders in clinical trials.

Previous studies have shown that patients like robots for care delivery.

AI to detect skin cancers

It was reported this past week that a new AI device is giving primary care doctors the ability to diagnose skin cancer, address access limitations in dermatology and avoid long wait times for dermatology referrals. 

Earlier this year, the US Food and Drug Administration authorized Miami-based DermaSensor, the first AI-enabled medical device for skin cancer detection in primary care. Reviewed under the De Novo pathway, the approval was considered a milestone, according to the story in Dermatology Times.

The device had a 96% sensitivity in detecting skin cancers in one study, Reuters said in January upon the FDA's approval.

While a supplemental study showed the device could assist in deciding lesion management, potentially reducing unnecessary referrals, a third clinical utility study showed a significant decrease in specificity for referrals. 

The device's efficacy across patient populations – particularly those historically underrepresented in clinical trials – is unclear, according to the story, and warrants ongoing post-market research to ensure equitable performance and monitor diagnostic accuracy.

Addressing staff turnover

To alleviate pressure on home-based healthcare providers, Griswold, one of the country’s largest home care franchises, with more than 170 locations in 32 states, and over 9,000 caregivers, reduced its caregiver turnover rate in its eight company-owned offices using AI for employee engagement.

The home care staff turnover rate is 77%, according to Ava cofounder and CEO Victor Hunt.

"Ava works to remove the burden of administrative work and reconnect caregivers and home care administrators with why they chose a healthcare career in the first place,” he said in an announcement with WellSky, a home health EHR vendor, last month.

But turnover isn't just about pay. Departing caregivers say they may have stayed with better engagement and communication, according to the EHR vendor. 

Ava's algorithms help predict caregiver turnover and proactively guide employee retention, reliability and recruitment through gamification, WellSky said. With the integration of an AI-driven employee engagement tool into its EHR, Griswold reported its caregiver turnover was reduced by 13% in seven months. 

"The staffing crisis facing our clients and the industry at large is a major concern, both for people seeking care in the home and for agencies trying to meet that need," Wes Little, chief analytics officer at WellSky, said in the statement.

Ava's communications also helped Griswold's caregivers clock in and out on time and meet minimum weekly hour requirements.

"We have seen tremendous benefit from using WellSky TeamEngage powered by Ava, both in the retainment of talented staff members and in the differentiation of our agency," Caitlin Griffith, Griswold's director of people and culture, said in a statement.

Andrea Fox is senior editor of Healthcare IT News.
Email: afox@himss.org
Healthcare IT News is a HIMSS Media publication.
 

The HIMSS AI in Healthcare Forum is scheduled to take place September 5-6 in Boston. Learn more and register.