Systems and methods for machine learning based automatic bullseye plot generation

A system, comprising:

• at least one storage device storing a set of instructions for generating a landmark detection network;
• and
• at least one processor configured to communicate with the at least one storage device, wherein when executing the set of instructions, the at least one processor is configured to direct the system to perform operations including:
• obtaining a plurality of training samples, each of the plurality of training samples including a sample slice image of a sample heart and a ground truth distance offset map of the sample slice image, the sample slice image being annotated with at least one ground truth landmark of the sample heart, and the ground truth distance offset map including distance information between each pixel and each of the at least one ground truth landmark of the sample slice image, the at least one ground truth landmark corresponding to a physical point that reflects an anatomical feature of at least one of a left ventricle or a right ventricle in the sample heart and being used for automatic bullseye plot generation;
• and
• generating the landmark detection network by training a preliminary model using the plurality of training samples.

A method for generating a bullseye plot of a heart of a subject, the method being implemented on a computing device having at least one processor and at least one storage device, the heart including a left ventricle, a right ventricle, and a myocardium around the left ventricle, the method comprising:

• obtaining multiple slice images in a plurality of groups, wherein the plurality of groups correspond to a plurality of sections substantially perpendicular to a long axis of the heart, each group corresponds to one of the plurality of sections and includes at least one slice image of the corresponding section, and each slice image includes part of the right ventricle, part of the left ventricle, and part of the myocardium;
• in each of the multiple slice images,
• identifying at least one landmark associated with the left ventricle by applying a landmark detection network, the at least one landmark corresponding to a physical point that reflects an anatomical feature of at least one of the left ventricle or the right ventricle and being used for automatic bullseye plot generation;
• and
• generating, based on the at least one landmark identified in each of the multiple slice images, the bullseye plot of the heart, wherein the bullseye plot includes a plurality of sectors, each of which represents an anatomical region of the myocardium in one of the plurality of sections.