Foundation Model based labelling of MR Shoulder images to drive Auto-Localizer workflow

Motivation: Develop automatic labelling capability on anatomical shoulder MRI images with minimal manual annotation. Goal(s): Leverage large-FOV, low resolution coil sensitivity maps to guide correct positioning of three-plane localizer for shoulder MRI planning. Approach: Use chained DINO-V2 and SAM foundation models, tuned to MRI localizers and a data driven similarity measure to label shoulder data at scale and transfer to low resolution coil sensitivity maps for CNN model training. Results: Excellent shoulder region localization with FM on anatomical (91% accuracy) and with CNN model on calibration data (error < 15 mm)Impact: A data adaptive, chained foundation model-based approach for annotating shoulder regions on MRI anatomical images at scale is shown. This allowed rapid development of model using low-resolution calibration data for correctly positioning three-plane localizer for shoulder anatomical planning and imaging.