Method and system for generating a medical image with localized artifacts using machine learning

A device for generating a medical image with localized artifacts, the device comprising:

• a processing unit;
• and
• a memory unit communicatively coupled to the processing unit, wherein the memory unit comprises an artifact recognition module configured to:
• receive a medical image generated by an imaging modality;
• identify a presence of one or more artifacts in the medical image using one or more trained machine learning models, wherein the one or more artifacts are one or more integral features in the medical image acquired from an anatomical region in a body of a patient, and wherein the identifying of the presence of the one or more artifacts in the medical image comprises:
• (1) identifying the one or more artifacts in the medical image using the one or more trained machine learning models, wherein each artifact of the one or more artifacts correspond to a predefined class of artifacts;
• and (2) determining a class associated with the one or more artifacts in the medical image using a predefined classification index;
• determine at least one artifact of interest from the one or more artifacts based on a predefined criteria;
• localize the at least one artifact of interest in the medical image using a post processing technique;
• and
• output the medical image comprising the at least one localized artifact of interest on a graphical user interface,
• wherein the localizing of the at least one artifact of interest comprises:
• determining data corresponding to a specific class of the at least one artifact of interest by back propagating the one or more trained machine learning models;
• computing discriminative regions indicative of a location of the at least one artifact of interest in the medical image based on the determined data corresponding to the specific class of the at least one artifact of interest;
• reconstructing an image comprising the discriminative regions indicative of the location of the at least one artifact of interest;
• and
• localizing the at least one artifact of interest in the medical image based on the reconstructed image.

A system comprising:

• at least one imaging modality for acquiring a medical image of a patient;
• and
• a device communicatively coupled to the at least one imaging modality, the device comprising a processing unit and a memory unit communicatively coupled to the processing unit, wherein the memory unit comprises an artifact recognition module configured to:
• receive a medical image generated by an imaging modality;
• identify a presence of one or more artifacts in the medical image using one or more trained machine learning models, wherein the one or more artifacts are one or more integral features in the medical image acquired from an anatomical region in a body of the patient, and wherein the identifying of the presence of the one or more artifacts in the medical image comprises:
• (1) identifying the one or more artifacts in the medical image using the one or more trained machine learning models, wherein each artifact of the one or more artifacts correspond to a predefined class of artifacts;
• and (2) determining a class associated with the one or more artifacts in the medical image using a predefined classification index;
• determine at least one artifact of interest from the one or more artifacts based on a predefined criteria;
• localize the at least one artifact of interest in the medical image using a post processing technique;
• and
• output the medical image comprising the at least one localized artifact of interest on a graphical user interface,
• wherein the localizing of the at least one artifact of interest comprises:
• determining data corresponding to a specific class of the at least one artifact of interest by back propagating the one or more trained machine learning models;
• computing discriminative regions indicative of a location of the at least one artifact of interest in the medical image based on the determined data corresponding to the specific class of the at least one artifact of interest;
• reconstructing an image comprising the discriminative regions indicative of the location of the at least one artifact of interest;
• and
• localizing the at least one artifact of interest in the medical image based on the reconstructed image.

A non-transitory computer readable storage medium comprising machine readable instructions, that when executed by a processing unit, cause the processing unit to:

• receive a medical image generated by an imaging modality;
• identify a presence of one or more artifacts in the medical image using one or more trained machine learning models, wherein the one or more artifacts are one or more integral features in the medical image acquired from an anatomical region in a body of a patient, and wherein the identifying of the presence of the one or more artifacts in the medical image comprises:
• (1) identifying the one or more artifacts in the medical image using the one or more trained machine learning models, wherein each artifact of the one or more artifacts correspond to a predefined class of artifacts;
• and (2) determining a class associated with the one or more artifacts in the medical image using a predefined classification index;
• determine at least one artifact of interest from the one or more artifacts based on a predefined criteria;
• localize the at least one artifact of interest in the medical image using a post processing technique;
• and
• output the medical image comprising the at least one localized artifact of interest on a graphical user interface,
• wherein the localizing of the at least one artifact of interest comprises:
• determining data corresponding to a specific class of the at least one artifact of interest by back propagating the one or more trained machine learning models;
• computing discriminative regions indicative of a location of the at least one artifact of interest in the medical image based on the determined data corresponding to the specific class of the at least one artifact of interest;
• reconstructing an image comprising the discriminative regions indicative of the location of the at least one artifact of interest;
• and
• localizing the at least one artifact of interest in the medical image based on the reconstructed image.