By: Jeff Deal, Conference Chair, Predictive Analytics World for Healthcare
In anticipation of his upcoming presentation at Predictive Analytics World for Healthcare Livestream, May 24-28, 2021, we asked Arjun Panesar, Founder and CEO of Diabetes Digital Media, a few questions about their deployment of predictive analytics. Catch a glimpse of his presentation, How big data and AI are being used to reverse type 2 diabetes and redefine “healthy”, and see what’s in store at the PAW Healthcare conference.
Q: In your work with predictive analytics, what behavior or outcome do your models predict?
A: DDM provides evidence-based health interventions (digital therapeutics) and health analytic AIs in a range of health domains, so we use predictive analytics in a range of settings. In terms of behaviour change which is core to user engagement, we use models to estimate the likelihood of patients to complete a particular program and level of intervention required. We use these analytics to best support healthcare professionals to support high-risk patients and minimise patient harm. It’s supporting the democratisation of healthcare, which is brilliant. In terms of deeper analytics, we provide live models that predict the trajectory of blood glucose, risk of hypoglycemia (low blood sugar levels), pancreatic cancer, and the risk of stress, anxiety and depression from interactions within our ecosystem that are combined with standardised questionnaires to provide an exceptionally accurate measure of mental health.
Q: How does predictive analytics deliver value at your organization – what is one specific way in which it actively drives decisions or operations?
A: Predictive analytics enables us to provide the epitome of evidence-based medicine – personalised evidence-based medicine to users. One AI we provide receives a longitudinal dataset and gives an indication of pancreatic cancer risk, which, as a condition, is most usually diagnosed when it’s too late. Our AI is able to use unstructured data and interactions, combined with blood glucose, nutrition and other health markers to provide a measure of pancreatic cancer risk. A key challenge in this area is generally the access to data and regulation on areas such as this.
Q: Can you describe a quantitative result, such as the predictive lift of your model or the ROI of an analytics initiative?
A: By deploying live AIs that measure levels of anger and depression within our community, we have been able to improve user satisfaction by 170%. By spotting users that may be exhibiting signs of mental discomfort, we are able to triage to appropriate health coach or healthcare professional which provides users a more comforting experience.
Q: What surprising discovery or insight have you unearthed in your data?
A: The novelty of longitudinal health data on non-Caucasian communities. Many AIs that are deployed in healthcare utilise datasets that do not represent well on non-white communities. Addressing these biases – particularly in health AI is crucial. A key strength is the diversity of our data which enables models to generalise well towards people of most backgrounds and cultures.
Q: Sneak preview: Please tell us a take-away that you will provide during your talk at Predictive Analytics World.
A: Gender and engagement within the first 7 days of joining a digital health intervention are key predictors of health outcomes. And no matter what your level of health, a deep recommendation engine is guaranteed to engage!
Arjun’s book, Machine Learning and AI in Healthcare is available of Amazon.
—————————–
Don’t miss Arjun’s presentation, How big data and AI are being used to reverse type 2 diabetes and redefine “healthy”, at PAW Healthcare on Monday, May 24, 2021 from 11:30 AM to 12:15 PM. Click here to register for attendance.
By: Jeff Deal, Conference Chair, Predictive Analytics World for Healthcare
 |
 |
 |
