Related Experiment Video
Updated: May 20, 2025
07:06
Destabilization of the Medial Meniscus and Cartilage Scratch Murine Model of Accelerated Osteoarthritis
Published on: July 6, 2022
4.2K
Using deep-learning based segmentation to enable spatial evaluation of knee osteoarthritis (SEKO) in rodent models
Jacob L Griffith1, Justin Joseph2, Andrew Jensen3
1J. Crayton Pruitt Family Department of Biomedical Engineering, University of Florida, Gainesville, FL, USA; Pain Research & Intervention Center of Excellence (PRICE), University of Florida, Gainesville, FL, USA.
Osteoarthritis and Cartilage
|March 26, 2025
Summary
A new deep learning tool, SEKO, quantifies and visualizes knee osteoarthritis (OA) joint remodeling in rodent models. This method accurately detects cartilage loss and bone changes, aiding OA research.
Area of Science:
- Biomedical Engineering
- Computational Biology
- Orthopedics
Background:
- Histology is crucial for evaluating joint remodeling in osteoarthritis (OA) preclinical models.
- Current histological analysis can be labor-intensive and subjective.
Purpose of the Study:
- Introduce a deep learning-driven histological analysis pipeline for spatial evaluation of knee osteoarthritis (SEKO).
- Quantify and visualize joint remodeling in the medial compartment of rodent knees.
Main Methods:
- Developed the SEKO pipeline with segmentation and visualization tools.
- Employed convolutional neural network architectures (HRNet, U-Net) for region identification.
- Calculated morphometric and location-dependent measures, generating probabilistic heat maps.
Main Results:
- Selected U-NET architecture for higher prediction accuracy.
- Identified cartilage loss comparable to manual segmentation.
- Detected subchondral bone area changes and location-dependent bone remodeling.
- Visualized spatial changes in cartilage thinning and bone remodeling via heat maps.
Conclusions:
- SEKO enables objective comparison of OA progression and therapeutic interventions.
- Provides visualization of spatial and morphometric changes in knee joint histology.
- Offers an open-source tool to advance collaborative OA research.
