Method, apparatus and system for automated spine labeling

A method for labeling one or more portions of a spine in an image of a human or animal body, the method comprising the steps of:

• a) matching a model of a spine segment with segments of the spine in the image by:
• starting matching the model of the spine segment with an initial segment of the spine in the image, wherein the initial segment of the spine in the image is located at an initial position along the spine in the image;
• and
• continuing matching the model of the spine segment with one or more further segments of the spine in the image, wherein the one or more further segments of the spine in the image are located at farther positions along the spine in the image, and the model of the spine segment relates to anatomical properties of one or more portions of the spine;
• and
• b) labeling the one or more portions of the spine in the image in response to step a);
• wherein
• an initial position of an initialization disk of the spine in the image is established by a disk profile corresponding to a string of region classes to which a set of disk candidates is mapped by classifying each disk candidate of the set of disk candidates to a region class or a region transition uncertainty;
• and
• the disk profile is matched to a full spine profile and multiple initialization disk candidates, which result from the region class or the region transition uncertainty in the disk profile, are resolved by repeating the labeling step.

An apparatus for labeling one or more portions of a spine in an image of a human or animal body, the apparatus comprising:

• an image processing unit configured or programmed to:
• a) match a model of a spine segment with segments of the spine in the image by:
• starting matching the model of the spine segment with an initial segment of the spine in the image, wherein the initial segment of the spine in the image is located at an initial position along the spine in the image;
• and
• continuing matching the model of the spine segment with one or more further segments of the spine in the image, wherein the further segments of the spine in the image are located at positions farther along the spine in the image, and the model of the spine segment relates to anatomical properties of one or more portions of the spine;
• and
• b) label the one or more portions of the spine in the image in response to step a), wherein
• the image processing unit is configured or programmed to establish an initial position of an initialization disk of the spine in the image by a disk profile corresponding to a string of region classes to which a set of disk candidates is mapped by classifying each disk candidate of the set of disk candidates to a region class or a region transition uncertainty;
• and
• the disk profile is matched to a full spine profile and multiple initialization disk candidates, which result from the region class or the region transition uncertainty in the disk profile, are resolved by repeating the labeling of the one or more portions of the spine in the image.