AI to Reduce Falls, Improve Outcomes, and Increase Revenues...Artificial Intelligence. Core...
Transcript of AI to Reduce Falls, Improve Outcomes, and Increase Revenues...Artificial Intelligence. Core...
AI to Reduce Falls,
Improve Outcomes,
and Increase Revenues
www.virtusense.ai
What if… you have
insights into the future and
time to act, then what would you do?
The use of mathematical procedures (algorithms) to analyze data.
Machine Learning
Object Recognition
Object Recognition
Object Recognition
Predictive Analytics
Source: Lab Tests Online
Predictive Model
ML
Preventive Action
Artificial Intelligence
Machine Learning
Artificial Intelligence vs Machine Learning
The replication of human analytical and/or decision-making capabilities
Artificial Intelligence
Core Components of AI
Data InputData (pre) Processing
Predictive Models
Decision Rules
Response / Output
What if…
you knew the probability
of each resident’s fall-risk in
the next year, week, or minute?
How would those insights
transform the way you provide care?
AI in Senior Living
Predict Fall-Risk 12 Months in Advance Mitigate Bedside Fall-Risk
Agenda
• Introduction (< 8 minutes)• Artificial Intelligence (AI)
• What if…
• Friendship Village: (20 minutes)• Proactive identification and care
• John Knox Village (20 minutes)• Continuously monitoring with alerts
• AI for Home Care (2 minutes)
• Q&A (10 minutes)
Friendship Village and VSTPresented by: Sarah Corbett
What if you knew every resident’s probability of falling in the next 12
months….. how would this change the way you provide care?
What if you could improve resident care while improving the top line?
• Resident care• Proactive identification and fall risk categorization
• Plan of Action Care based on mobility ranking (wellness, therapy, etc.)• Classes: Balance, Strength, Tai Chi, Silver Sneakers, Parkinson’s, etc.
• Top Line• Med B Therapy referrals generate approximately $1,300 in revenue for
Friendship Village
• Wellness Program increase by about 10% (specifically the balance class)• Residents that participate in wellness generally stay in IL longer- higher IL occupancy
means more revenue
• Market Differentiation
Population Health Dashboard
Of the 63 High-Risk Residents…
What if you could monitor residents’ fall risks and be proactive to catch health declines?
• Gait Analysis testing gives a fall risk percentage based on the likeliness of a resident 70+ to fall in the next 12 months
• Retest every 6 months to monitor how the action plan is going, if there have been major declines in health, what the next steps are.
• Retesting allows Friendship Village to continually engage with the residents and helps us reach people who don’t normally seek out help
• Log into the web dashboard to compare all of their previous tests to monitor their balance changes over a long period of time.
What if your wellness program could grow from the use of AI?
• Quantifying a resident’s fall risk helps them see the reality of their balance• Creates a sense of urgency to begin a wellness program
• Wellness Classes have grown
• Fitness Center usage has increased
• Positive view of therapy
Success Stories
• Mary was a low mobility resident due to her bad neuropathy. After taking the assessment, she asked her doctor for orders for therapy and her doctor didn’t give her the orders. She ended up switching doctors to get the help she needed
• Claude was a low mobility resident and couldn’t even complete some of the tests. He went to therapy and worked in the pool. We retested him and his fall risk improved by 9%!
• Marcella is a 95 year-old resident who is very active and ranked high mobility. She asked for a print-out of her results to brag to her family about how the wellness programs are helping her.
• Pat has had 2 falls in her apartment over the last year and couldn’t figure out why. After using VST to test her, we discovered it was the neuropathy causing her to lose her balance. She was referred to therapy for treatment.
What if…
• You could predict a fall?
• You could predict a decline?
• You could prevent a fall?
• You could ensure residents aged in place?
The Backstory
• Fall rate for the Village Rehab Unit
• Average daily census of 52 for the past 18 months
• Increased acuity, increased falls
VSTAlert Data (Jul 6 – Aug 23)
• Total Patient Days: 328
• # Falls: 0* when VSTAlert was on with a clear field of view• One fall occurred when VSTAlert was turned off• One fall occurred when a curtain was covering VSTAlert
Alarms per Patient Day
• Bed Exits 5.6
• 2nd Person 2.6
• False Alarms 1.7
• Other 0.4
Potential fall-risk
Nurse forgot to pause VSTAlert upon entry. Developing an auto-pause feature for this event.
The Reception
• Families
• Associates
• The Village
The Future
• What if we could alter the care delivery system?
• What if we could use AI to lengthen AND enhance our resident’s lives?
AI for Home Care
Questions?
Sarah Corbett, Friendship Village - Sunset Hills
Anthony Columbatto, John Knox Village
David Park, VirtuSense Technologies