Lead University: Carnegie Mellon University
PI: Jonathan Cagan, Mechanical Engineering
Co-PIs: Phil LeDuc, Mechanical Engineering; Sam Akhavan, Allegheny Health Network
Labral tears affect a large number of patients playing sports reliant on upper body strength and are present in 23% of a randomly sampled population. In cases, these patients are found to have instability in the tear requiring surgical intervention. Currently, diagnosis of labral tears is achieved using Magnetic Resonance (MR) Arthrography, relying on contrast to highlight the separation between the glenoid and labrum, and extensive assessment from radiologists. MR Arthrograms are common and safe but utilize an added procedure where dye is injected into the shoulder to delineate the tear. This can be time consuming, uncomfortable for patients, and more costly. In addition, patients may have to be followed over time to track the evolution of a tear.
If diagnosis of labral tears could be made reliably using standard MRI coupled with predictive indicators based on finite element analysis, patients would benefit significantly and the total cost of care would decrease. We believe that through using computational approaches and creating morphological shape grammar analysis of standard MRIs to diagnose labral tears, and predictive modeling, we can improve the accuracy of unenhanced MRIs in diagnosing this pathology. Part of the solution will be sufficient modeling of the complex interaction between soft and hard tissues that will enable better diagnostics based upon MRI data; we will partner with ANSYS, a PA company, to enable an effective 3D analysis of cartilaginous deformation in tear and no tear cases.
Providing additional information or insight to radiologists could result in fewer false negatives, providing more accurate diagnosis and resulting in more effective and timely treatment. Furthermore, this approach could potentially enable less invasive MRI imaging to be used to predict the tear or decrease the need for the more invasive MR Arthrograms, which would simplify patient procedures significantly and reduce the cost of diagnosis.