Can Early-Stage Risk Assessment Tools be the Game-changer in Kidney Care?
A quick Q&A with UCLA’s Dr. Keith Norris.
By Elise Wilfinger
I watched a YouTube video (UCLA 8th Annual CORE Kidney Health Conference 2022, March 11, 2022) featuring one of your UCLA colleagues, Dr. Naveen Raja, Medical Director of Population Health. His remarks were focused on the importance of early stage diagnosis. He called it ‘the critical pillar,’ the need to catch kidney function decline early enough so that doctors had the best chance of slowing the progression.
He noted that UCLA was in the process of creating an AI-machine learning model that “predicts how soon a patient’s kidneys may worsen…by knowing (this information) we can maximize our efforts in slowing that progression.” What’s your take on a model of this kind?
Naveen works closely with our colleagues at the UCLA CORE Kidney Health Program to best determine how to impact care in earlier stages. One of their initiatives is to figure out how to make available data that much more actionable for doctors and their patients. That’s where the machine learning model comes into play; I believe they are looking into two different options:
- Directing the computer algorithm to look at specific variables when creating the model
- Directing the computer to look at all variables to calculate which are the most relevant for the model
The goal is to have an AI model – inclusive of kidney biomarkers and EHR data – that would find predictive patterns much more quickly and accurately (than humans). In this, the provider would have UACR, eGFR (traditional metrics) and a risk prediction model score at his/her fingertips enabling better, more comprehensive decision making and outcomes for each patient.
Join the List
To be one of the first to read our upcoming blogs and other content, subscribe to Well FilteredTM here.
Spread the word
If you enjoyed this post, we’d be grateful if you would help us spread its content.