Skip to main content
Fig. 6 | BMC Veterinary Research

Fig. 6

From: A method for labelling lesions for machine learning and some new observations on osteochondrosis in computed tomographic scans of four pig joints

Fig. 6

Preparation of CT scans for labelling. a The original scan (green box) was treated as an array of ~ 1000 slices with 512 × 512 pixels. b-c A new scan volume (blue box) was constructed around the femur based on image moment invariants. The grid size for the new femur scan volume was 300 × 150 × 150 voxels. c-d The distal end of the new volume was used to extract a volume containing the stifle joint, where transverse plane images were perpendicular to the long axis of the femur, irrespective of limb position in the original scan (a). d-e The x, y and z coordinates from the original CT scan remained associated with the new voxels, meaning applied labels (red areas) could be translated back to their location within the original scan

Back to article page