# US 2021/0279891 A1 · Detecting subject motion in medical imaging > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0279891 A1 - Document ID: US20210279891A1 - Status: Published application - Kind: A1 - Application number: 17/259,208 - Filed: 2019-07-11 - Published: 2021-09-09 - Earliest claimed: 2018-07-13 - Assignees: Koninklijke Philips N.V. - Applicants: Koninklijke Philips N.V. - Inventors: Dinesh MYSORE SIDDU; Krishnamoorthy PALANISAMY; Christopher Stephen HALL; Nagaraju BUSSA - Web page: https://patentgrep.com/patent/US20210279891A1 ## CPC - G06T 7/246: using feature-based methods, e.g. the tracking of corners or segments - G06T 7/194: involving foreground-background segmentation - G06T 2207/10081: Computed x-ray tomography [CT] - G06T 2207/10088: Magnetic resonance imaging [MRI] - G06T 2207/10104: Positron emission tomography [PET] - G06T 2207/20081: Training; Learning - G06T 2207/20084: Artificial neural networks [ANN] - G06T 2207/20124: Active shape model [ASM] - G06T 2207/30004: Biomedical image processing - G06T 2207/30016: Brain - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL ## Abstract Presented are concepts for detecting subject motion in medical imaging of a subject. One such concept obtains a motion classification model representing relationships between motion of image features and subject motion values. For each of a plurality of medical slice images of an imaged volume of the subject, an image feature of the medical slice image is extracted. Based on the extracted image feature for each of the plurality of medical slice images, motion information for the image feature is determined. Based on the motion information for the image feature and the obtained motion classification model, a subject motion value is determined. ## Claims ### Claim 1 A method for detecting subject motion in medical imaging of a subject, the method comprising: obtaining a motion classification model representing relationships between motion of image features and subject motion values; for each of a plurality of medical slice images of an imaged volume of the subject, extracting an image feature of the medical slice image; based on the extracted image feature for each of the plurality of medical slice images, determining motion information for the image feature, the motion information representing motion of the image feature in the imaged volume of the subject; and determining a subject motion value based on the motion information for the image feature and the obtained motion classification model. ### Claim 2 (depends on 1) The method of claim 1, wherein obtaining a motion classification model comprises: generating a motion classification model based on historical data relating to previously determined subject motion values for image features extracted from medical slice images. ### Claim 3 (depends on 2) The method of claim 2, wherein obtaining a motion classification model further comprises: refining the motion classification model using a machine learning algorithm with training data relating to subject motion values for image features of medical slice images. ### Claim 4 (depends on 1) The method of claim 1, wherein the step of extracting an image feature of a medical slice image comprises: separating the medical slice image into a foreground and a background; and extracting a foreground image feature from the foreground. ### Claim 5 (depends on 4) The method of claim 4, wherein extracting a foreground image feature comprises identifying an image feature from at least one of: a spatial domain representation of the foreground; a wavelet domain representation of the foreground; and a spectral domain representation of the foreground; and extracting the identified image feature as the foreground image feature. ### Claim 6 (depends on 1) The method of claim 1 wherein the step extracting an image feature of a medical slice image comprises: segregating the medical slice image into a foreground and a background; and extracting a background image feature from the background. ### Claim 7 (depends on 6) The method of claim 6, wherein extracting a background image feature comprises: applying radon transformations to the background to generate a plurality of radon transform profiles; identifying an image feature based on the radon transform profiles; and extracting the identified image feature as the background image feature. ### Claim 8 (depends on 1) The method of claim 1, further comprising: based on the motion information for the image feature, identifying a medical slice image in which motion of the extracted image feature exceeds a threshold value. ### Claim 9 (depends on 8) The method of claim 8, further comprising: determining the threshold value (Th) based on historical data relating to previously determined motion of the image feature. ### Claim 10 (depends on 1) The method of claim 1, wherein the motion classification model represents relationships between motion of image features and subject motion values for a plurality of different classification methods, and wherein determining a subject motion value comprises: selecting relationships between motion of image features and subject motion values for one of the plurality of different classification method based on the extracted image feature; and determining a subject motion value based on the motion information for the image feature and the selected relationships between motion of image features and subject motion values. ### Claim 11 (depends on 1) A computer program product downloadable from a communications network and/or stored on a computer readable medium and/or microprocessor-executable medium wherein the computer program product comprises computer program code instructions, which when executed by at least one processor, implement a method as claimed in claim 1. ### Claim 12 (depends on 12) A system comprising at least one processor and the computer program product of claim 12. ### Claim 13 A system for detecting subject motion in medical imaging of a subject, the system comprising: an interface component adapted to obtain a motion classification model representing relationships between motion of image features and subject motion values; a feature extraction component adapted, for each of a plurality of medical slice images of an imaged volume of the subject, to extract an image feature of the medical slice image; a data processing component adapted to determine motion information for the image feature based on the extracted image feature for each of the plurality of medical slice images, the motion information representing motion of the image feature in the imaged volume of the subject; and a motion determination component adapted to determine a subject motion value based on the motion information for the image feature and the obtained motion classification model. ### Claim 14 (depends on 13) The system of claim 13, wherein the interface component is adapted to generate a motion classification model based on historical data relating to previously determined subject motion values for image features extracted from medical slice images. ### Claim 15 (depends on 13) The system of claim 13, wherein the feature extraction component is adapted to separate the medical slice image into a foreground and a background, and to extract at least one of: a foreground image feature from the foreground; and a background image feature from the background.