Using machine learning for feature detection and extraction from MRI

Project Description

Training machine learning models for the extraction of several features from different MRI sequences…

From PD sequences, we can detect/extract bone, cartilage, and meniscus tears. From T2 sequences, we can detect/extract bone, bone marrow lesions (BMLs), and cartilage. This project would build on our previous work:

Stirling CE, Neeteson NJ, Walker REA, Boyd SK, 2024. Deep learning-based automated detection and segmentation of bone and traumatic bone marrow lesions from MRI following an acute ACL tear. Computers in Biology and Medicine 178, 108791.

A model to segment the meniscus could be developed and then compared to our surgical reports from the SALTAC and SALTACII studies.

Scope

• PhD/MSc side project
• Summer student

Data Source

• SALTAC
• SALTACII

Resources

• Python scripts
• ML computer