NA-MIC Project Weeks
Back to Projects List
Shape Variation Analyzer
Key Investigators
- Juan-Carlos Prieto (University of North Carolina)
- Nina Tubau (University of Michigan)
Project Description
Osteoarthritis (OA), the most prevalent arthritis worldwide, is associated with significant pain and disability. The complex pathogenesis of temporomandibular joint (TMJ) OA remains unclear to this day, and its course challenges experts given the different morphological patterns of bone resoption and formation observed in its various stages. We developed ShapeVariationAnalyzer (SVA), a neural network to classify morphological variations using 3D models of the mandibular condyle that can provide information about bony changes and disease changing in TMJ OA.
Objective
- Develop a noninvasive technique to provide information about bony changes and disease changing in TMJ OA
- Classify morphological variations using 3D models of the mandibular condyle
- Build an accurate neural network for this classification
Approach and Plan
- Extract other features to run the neural network and make it more accurate
- Test with this new model
Progress and Next Steps
- Data generated for each group with SMOTE algorithm
- Improvement of the neural network
- Testing the accuracy
- Future: Add biological and clinical data to the neural network
Illustrations
