What is it about?
Chronic kidney disease often has no early symptoms, leaving many people undiagnosed. We developed MERWACS, an artificial intelligence tool that identifies adults who should be screened for kidney disease. Instead of requiring blood or urine tests, it uses up to 12 simple, non-invasive measurements like age, weight, and blood pressure. We trained the model using 30 years of U.S. health data and successfully validated it on an independent dataset from South Korea.
Featured Image
Photo by Robina Weermeijer on Unsplash
Why is it important?
Early detection of kidney disease is crucial to slow its progression, but running laboratory tests on the entire population is not realistic. MERWACS provides an accessible early warning system to help prioritize exactly who needs a definitive lab test. Uniquely, our tool accounts for the natural decline in kidney function that occurs with healthy aging. This ensures we do not unnecessarily alarm healthy older adults while still identifying true pathological disease.
Perspectives
As a data scientist, my focus is translating massive datasets into practical solutions that benefit society. It was fascinating to discover that simple physical traits can assess metabolic risk almost as effectively as complex laboratory data. By intentionally excluding blood tests, we sacrificed a small amount of diagnostic precision to achieve massive public accessibility. I hope this framework inspires new ways to approach early detection for other silent diseases.
Daniel Yoo
Danmarks Tekniske Universitet
Read the Original
This page is a summary of: MERWACS: Development and external validation of a non-invasive machine learning tool for identifying subjects to be screened for CKD, PLOS Digital Health, July 2026, PLOS,
DOI: 10.1371/journal.pdig.0001486.
You can read the full text:
Resources
Contributors
The following have contributed to this page
