PIPages
Labs

Xiaoxiao Sun

University of Arizona, AZ·Mathematics & Statistics

Knee osteoarthritis, a wearing-down of joint tissues, is the most common musculoskeletal disorder and affects more than 300 million people worldwide. Knee pain from osteoarthritis is a leading cause of physical disability and hospital visits.

Fields

Mathematics & Statistics

Computer Science

Medicine & Clinical Research

Topics

knee osteoarthritis

knee pain

pain spatial patterns

Grants
1 known grant, through Aug 2027

What they work on

  1. Adapt a Transformer-based deep learning model, a neural network suited to complex patterns, to predict new localized, regional or diffuse knee pain from knee MRIs in the Osteoarthritis Initiative cohort.
  2. Develop an interpretable statistical framework that describes how bone marrow lesions, abnormal areas of bone seen on MRI, relate to new localized, regional or diffuse knee pain.
  3. Test whether structural abnormalities on knee MRI early in the disease course determine which spatial pattern of knee pain a person develops.

More about the research

Pain can be localized to one spot, spread across a region, or diffuse across the whole knee, but past studies have not systematically accounted for these patterns. Existing prediction tools can be accurate yet hard to interpret, so they reveal little about why pain arises.

Why it matters

Tools that flag knees at high risk of new pain in specific locations, and the MRI features behind that risk, could help select patients most likely to benefit from drug and non-drug treatments in clinical trials aimed at preventing or slowing knee osteoarthritis. This may help reduce the public health burden of the disease.

Known grants

From NIH and NSF research grants. Labs often have other funding we can't see.

Computational Tools for Predicting and Understanding Knee Pain Spatial Patterns

NIH · Sep 2024 – Aug 2027

Closest labs

David Felson

Boston University, MA·Medicine & Clinical Research

Identify proteins in knee synovial fluid and plasma linked to knee pain and MRI-detected cartilage damage, using SOMAScan aptamer-based proteomics in about 500 paired samples from the MOST cohort.

Lakshmi Nair

University of Connecticut, CT·Biomedical Engineering

Optimize where around the knee a new local anesthetic formulation is injected (periarticular placement), testing injection locations to find the one that gives the longest and strongest relief of osteoarthritis pain.

Patricia Franklin

Northwestern University, IL·Medicine & Clinical Research

Advance a genetics-enabled learning health system at Northwestern that uses electronic health record genetic results, patient-reported data and web tools to bring emerging genetic evidence into hospital and community clinics.

The school

University of Arizona: 36 labs on PI Pages, $1.0B a year on research, 487 research doctorates a year.

More about University of Arizona

Reach out

A start for your email. Fill in the brackets.