Systems and methods for extracting vertebral bodies from two-dimensional image data and classifying fracture

At least one computer readable media encoded with instructions, which, when executed, cause a system to perform operations comprising:

• extract, using a first trained machine learning model, image data portions corresponding to respective individual vertebral bodies from a two-dimensional image of at least a portion of a spine;
• input each of the image data portions to a second machine learning model to generate respective scores indicative of a likelihood each of the individual vertebral bodies has a fracture;
• and
• classify the spine, using a subject-level classifier based on the scores, into a fracture status.

A system comprising:

• at least one processor;
• at least one computer readable media encoded with instructions which, when executed by the at least one processor, cause the system to perform operations comprising:
• extract, using a first trained machine learning model, image data portions corresponding to respective individual vertebral bodies from a two-dimensional image of at least a portion of a spine;
• input each of the image data portions to a second machine learning model to generate respective scores indicative of a likelihood each of the individual vertebral bodies has a fracture;
• and
• classify the spine, using a subject-level classifier based on the scores, into a fracture status.