CTNF 18/883,814 CTNF 88525 2667 DETAILED ACTION Claim Rejections - 35 USC § 103 07-20-aia AIA The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action: A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made. 07-21-aia AIA Claim s 1-5, and 13-16 are rejected under 35 U.S.C. 103 as being unpatentable over KARANAM (US 20210158932 A1), in view of Liu (US 12118766 B2, Date Filed 2021-10-22) . Re Claim 1, KARANAM discloses a keypoint detection method for medical imaging (see KARANAM: e. g., --the processor 202 may be configured to receive an image of the patient from the sensing device and extract a set of features from the image that collectively represent a characteristic of the patient. The processor 202 may be further configured to match at least one of the extracted features against the imagery data (e.g., known features of the patient) stored in the feature database 212 to determine an identity of the patient. [0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like.--, in [0027]-[0028]), characterized by comprising: receiving an image sequence, the image sequence comprising a plurality of images of an object (see KARANAM: e.g., -- scan images previously taken for the patient, positions of the patient during those scans, adjustments or corrections made to get the patient into a desired scan position, overlaid scan images and pictures (or models) of the patient , etc. The personalized medical assistance information may be displayed in various formats including, for example, videos, animations, and/or AR presentations . For example, the overlaid representations of the patient's scan images and pictures may be displayed in an AR environment in which a physician equipped with AR glasses and/or an AR input device may swipe through the representations in a stereoscopic manner.--, in [0024]); performing keypoint detection on the plurality of images separately (see KARANAM: e. g., --the processor 202 may be configured to receive an image of the patient from the sensing device and extract a set of features from the image that collectively represent a characteristic of the patient. The processor 202 may be further configured to match at least one of the extracted features against the imagery data (e.g., known features of the patient) stored in the feature database 212 to determine an identity of the patient [0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like.--, in [0027]-[0028]); KARANAM however does not explicitly disclose generating a keypoint image sequence; LIU discloses generating a keypoint image sequence (see LIU: e.g., -- the 3D pose data may be obtained by fusing the 2D pose data and depth data. The 2D pose data are 2D coordinates of body keypoints in an RGB image, and the 3D pose data are 3D body keypoints. By means of the 3D pose data, detection of keypoints of a human body may be performed accurately when the human body is in a motion state. For example, a motion state is decomposed into at least one node pose of raising a hand, kicking a leg, shaking a head or bending down, so that body keypoint detection corresponding to these node poses is tracked in real time.--, in lines 15-25, col. 4; also see: -- . For the first type of data (RGB data), after the first image processing, whether a human body is detected in the present image frame is determined by using a trained body tracking network; and if the body is detected, corresponding target RGB data in the present image frame are processed in a subsequent step. In the subsequent step, for the second type of data (RGBD data), after the second image processing, the RGBD data are fused with the target RGB data to obtain 3D pose data (3D coordinates of body skeleton keypoints) according to the RGBD data and the target RGB data. Dynamic tracking includes: using the 3D coordinates to represent the 3D pose data of the body skeleton keypoints to implement tracking when the body is in the motion state, such as tracking a change of node pose pairs, which supports at least one body action of raising the hand, kicking the leg, shaking the head, bending down, etc ….. (32) FIG. 3 illustrates a schematic diagram of body skeleton keypoints according to an embodiment of the present disclosure. The body skeleton includes the number of 17 keypoints; and by detection of the 17 keypoints, a dynamic pose change of the user , such as at least one body action of raising the hand, kicking the leg, shaking the head, bending down and the like, may be tracked in real time. 33) For instance, in a scenario on how the user holding the mobile phone terminal interacts with a large-screen device such as a television, a first body motion state (such as a swing action when the user plays tennis) is acquired; and a change of body keypoint positions corresponding to the first body motion state is described through first 3D pose data. --, in line 30, col 5 through line 17, col. 6); KARANAM and LIU are combinable as they are in the same field of endeavor: monitoring and detection of keypoints of a patient/human body. Therefore it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to further modify KARANAM’s method using LIU’s teachings by including generating a keypoint image sequence to KARANAM’s image processing and keypoints detection to track a dynamic pose change of the user (see LIU: e.g., in lines 15-25, col. 4; and in line 30, col 5 through line 17, col. 6); KARANAM as modified by LIU further disclose performing keypoint occlusion detection on the images in the image sequence (see KARANAM: e.g., -- [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0029]-[0030]); determining keypoint distribution information according to the keypoint image sequence and a result of the keypoint occlusion detection (see KARANAM: e.g., --0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033] {above “keypoints positions” read on claimed limitation of “keypoint distribution information”}; also see LIU: e.g., --a first body motion state (such as a swing action when the user plays tennis) is acquired; and a change of body keypoint positions corresponding to the first body motion state is described through first 3D pose data…. a second body motion state (for example, the user tries to raise both hands to form 90° with the horizontal plane) is acquired, and the change of body keypoint positions corresponding to the second body motion state is described through second 3D pose data. ….a third body motion state (such as a pose that the user runs) is acquired, and a change of body keypoint positions corresponding to the third body motion state is described through third 3D pose data. --, in line 12, col. 6, through line 14, col. 7). Re Claim 2, KARANAM as modified by LIU further disclose the image sequence comprises at least one first image, keypoints in at least a partial region of the first image being unoccluded (see KARANAM: e. g., --Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0029]-[0030]). Re Claim 3, KARANAM as modified by LIU further disclose the image sequence comprises a color image sequence and a depth image sequence corresponding to each other, and the keypoint detection comprises performing keypoint detection on a color image in the color image sequence and a depth image in the depth image sequence (see KARANAM: e.g., --Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more image s of the patient may be used to determine the obstructed or occluded areas. The processor 202 may be also configured to use features associated with an obstructed area for patient matching and provide an indication that such matching may not be robust (e.g., give the matching a low confidence score).--, in [0030]; also see LIU: e.g., -- each data pair formed by a RGB and a RGBD is an image frame corresponding to the same view. Body keypoints of each frame of image in the RGB image data stream are aligned to depth data of body keypoints corresponding to the same image, so that for any one body keypoint in the image, both a 2D coordinate for characterizing a position of the body keypoint and a distance value for characterizing a change of movement of the body keypoint are obtained, and thus 3D information for the body keypoint position is obtained. The depth data are acquired from a DepthMap. In an acquisition scene including a camera or a camera module, the DepthMap may be considered as: an image (or referred as an image channel) that includes information on the distance to a surface of a target object acquired in the scene . When the distance of at least one point relative to the camera or the camera module in the scene is represented by a DepthMap, each pixel value in the DepthMap may represent a distance between a point and the camera in the scene.--, in line 57, col. 4, through line 8, col. 5; and, -- the 3D pose data may be obtained by fusing the 2D pose data and depth data . The 2D pose data are 2D coordinates of body keypoints in an RGB image, and the 3D pose data are 3D body keypoints. By means of the 3D pose data, detection of keypoints of a human body may be performed accurately when the human body is in a motion state. For example, a motion state is decomposed into at least one node pose of raising a hand, kicking a leg, shaking a head or bending down, so that body keypoint detection corresponding to these node poses is tracked in real time.--, in lines 15-25, col. 4; also see: -- . For the first type of data (RGB data), after the first image processing, whether a human body is detected in the present image frame is determined by using a trained body tracking network; and if the body is detected, corresponding target RGB data in the present image frame are processed in a subsequent step. In the subsequent step, for the second type of data (RGBD data), after the second image processing, the RGBD data are fused with the target RGB data to obtain 3D pose data (3D coordinates of body skeleton keypoints) according to the RGBD data and the target RGB data. Dynamic tracking includes: using the 3D coordinates to represent the 3D pose data of the body skeleton keypoints to implement tracking when the body is in the motion state, such as tracking a change of node pose pairs, which supports at least one body action of raising the hand, kicking the leg, shaking the head, bending down, etc ….. (32) FIG. 3 illustrates a schematic diagram of body skeleton keypoints according to an embodiment of the present disclosure. The body skeleton includes the number of 17 keypoints; and by detection of the 17 keypoints, a dynamic pose change of the user , such as at least one body action of raising the hand, kicking the leg, shaking the head, bending down and the like, may be tracked in real time. 33) For instance, in a scenario on how the user holding the mobile phone terminal interacts with a large-screen device such as a television, a first body motion state (such as a swing action when the user plays tennis) is acquired; and a change of body keypoint positions corresponding to the first body motion state is described through first 3D pose data. --, in line 30, col 5 through line 17, col. 6). Re Claim 4, KARANAM as modified by LIU further disclose the color image sequence comprises at least one first color image, keypoints in at least a partial region of the first color image being unoccluded (see LIU: e.g., -- the 3D pose data may be obtained by fusing the 2D pose data and depth data . The 2D pose data are 2D coordinates of body keypoints in an RGB image, and the 3D pose data are 3D body keypoints. By means of the 3D pose data, detection of keypoints of a human body may be performed accurately when the human body is in a motion state. For example, a motion state is decomposed into at least one node pose of raising a hand, kicking a leg, shaking a head or bending down, so that body keypoint detection corresponding to these node poses is tracked in real time.--, in lines 15-25, col. 4; also see: -- . For the first type of data (RGB data), after the first image processing, whether a human body is detected in the present image frame is determined by using a trained body tracking network; and if the body is detected, corresponding target RGB data in the present image frame are processed in a subsequent step. In the subsequent step, for the second type of data (RGBD data), after the second image processing, the RGBD data are fused with the target RGB data to obtain 3D pose data (3D coordinates of body skeleton keypoints) according to the RGBD data and the target RGB data. Dynamic tracking includes: using the 3D coordinates to represent the 3D pose data of the body skeleton keypoints to implement tracking when the body is in the motion state, such as tracking a change of node pose pairs, which supports at least one body action of raising the hand, kicking the leg, shaking the head, bending down, etc ….. (32) FIG. 3 illustrates a schematic diagram of body skeleton keypoints according to an embodiment of the present disclosure. The body skeleton includes the number of 17 keypoints; and by detection of the 17 keypoints, a dynamic pose change of the user , such as at least one body action of raising the hand, kicking the leg, shaking the head, bending down and the like, may be tracked in real time. 33) For instance, in a scenario on how the user holding the mobile phone terminal interacts with a large-screen device such as a television, a first body motion state (such as a swing action when the user plays tennis) is acquired; and a change of body keypoint positions corresponding to the first body motion state is described through first 3D pose data. --, in line 30, col 5 through line 17, col. 6). Re Claim 5, KARANAM as modified by LIU further disclose the depth image sequence comprises at least one first depth image, keypoints in at least a partial region of the first depth image being unoccluded (see LIU: e.g., -- each data pair formed by a RGB and a RGBD is an image frame corresponding to the same view. Body keypoints of each frame of image in the RGB image data stream are aligned to depth data of body keypoints corresponding to the same image, so that for any one body keypoint in the image, both a 2D coordinate for characterizing a position of the body keypoint and a distance value for characterizing a change of movement of the body keypoint are obtained, and thus 3D information for the body keypoint position is obtained. The depth data are acquired from a DepthMap. In an acquisition scene including a camera or a camera module, the DepthMap may be considered as: an image (or referred as an image channel) that includes information on the distance to a surface of a target object acquired in the scene. When the distance of at least one point relative to the camera or the camera module in the scene is represented by a DepthMap, each pixel value in the DepthMap may represent a distance between a point and the camera in the scene .--, in line 57, col. 4, through line 8, col. 5; ; also see: -- the 3D pose data may be obtained by fusing the 2D pose data and depth data . The 2D pose data are 2D coordinates of body keypoints in an RGB image, and the 3D pose data are 3D body keypoints. By means of the 3D pose data, detection of keypoints of a human body may be performed accurately when the human body is in a motion state. For example, a motion state is decomposed into at least one node pose of raising a hand, kicking a leg, shaking a head or bending down, so that body keypoint detection corresponding to these node poses is tracked in real time.--, in lines 15-25, col. 4; also see: -- . For the first type of data (RGB data), after the first image processing, whether a human body is detected in the present image frame is determined by using a trained body tracking network; and if the body is detected, corresponding target RGB data in the present image frame are processed in a subsequent step. In the subsequent step, for the second type of data (RGBD data), after the second image processing, the RGBD data are fused with the target RGB data to obtain 3D pose data (3D coordinates of body skeleton keypoints) according to the RGBD data and the target RGB data. Dynamic tracking includes: using the 3D coordinates to represent the 3D pose data of the body skeleton keypoints to implement tracking when the body is in the motion state, such as tracking a change of node pose pairs, which supports at least one body action of raising the hand, kicking the leg, shaking the head, bending down, etc ….. (32) FIG. 3 illustrates a schematic diagram of body skeleton keypoints according to an embodiment of the present disclosure. The body skeleton includes the number of 17 keypoints; and by detection of the 17 keypoints, a dynamic pose change of the user , such as at least one body action of raising the hand, kicking the leg, shaking the head, bending down and the like, may be tracked in real time. 33) For instance, in a scenario on how the user holding the mobile phone terminal interacts with a large-screen device such as a television, a first body motion state (such as a swing action when the user plays tennis) is acquired; and a change of body keypoint positions corresponding to the first body motion state is described through first 3D pose data. --, in line 30, col 5 through line 17, col. 6). Re Claim 13, KARANAM as modified by LIU further disclose determining keypoint distribution information on the basis of the method according to claim 1 (see KARANAM: e.g., --0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033]); and performing a scanning operation according to the determined keypoint distribution information (see KARANAM: e.g., --0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: -- scan images previously taken for the patient, positions of the patient during those scans, adjustments or corrections made to get the patient into a desired scan position, overlaid scan images and pictures (or models) of the patient , etc. The personalized medical assistance information may be displayed in various formats including, for example, videos, animations, and/or AR presentations . For example, the overlaid representations of the patient's scan images and pictures may be displayed in an AR environment in which a physician equipped with AR glasses and/or an AR input device may swipe through the representations in a stereoscopic manner.--, in [0024]; and, --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033]). Re Claim 14, KARANAM as modified by LIU further disclose wherein performing the scanning operation according to the determined keypoint distribution information comprises: positioning the scan object according to the keypoint distribution information (see KARANAM: e.g., --0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: -- scan images previously taken for the patient, positions of the patient during those scans, adjustments or corrections made to get the patient into a desired scan position, overlaid scan images and pictures (or models) of the patient , etc. The personalized medical assistance information may be displayed in various formats including, for example, videos, animations, and/or AR presentations . For example, the overlaid representations of the patient's scan images and pictures may be displayed in an AR environment in which a physician equipped with AR glasses and/or an AR input device may swipe through the representations in a stereoscopic manner.--, in [0024]; and, --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033]). Re Claim 15, claim 15 is the corresponding system claim to claim 1 respectively. Claim 15 thus is rejected for the similar reasons for claim 1. See above discussions with regard to claim 1 respectively. KARANAM as modified by LIU further disclose medical imaging system, comprising: a controller, configured to perform the keypoint detection method for medical imaging according to claim 1 (see KARANAM: e.g., Fig. 1, and --[0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: -- scan images previously taken for the patient, positions of the patient during those scans, adjustments or corrections made to get the patient into a desired scan position, overlaid scan images and pictures (or models) of the patient , etc. The personalized medical assistance information may be displayed in various formats including, for example, videos, animations, and/or AR presentations . For example, the overlaid representations of the patient's scan images and pictures may be displayed in an AR environment in which a physician equipped with AR glasses and/or an AR input device may swipe through the representations in a stereoscopic manner.--, in [0024]; and, --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033]); and a scanning assembly, performing a scanning operation according to keypoint distribution information determined by the controller (see KARANAM: e.g., Fig. 1, and --[0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: --scan images previously taken for the patient, positions of the patient during those scans, adjustments or corrections made to get the patient into a desired scan position, overlaid scan images and pictures (or models) of the patient, etc. The personalized medical assistance information may be displayed in various formats including, for example, videos, animations, and/or AR presentations. For example, the overlaid representations of the patient's scan images and pictures may be displayed in an AR environment in which a physician equipped with AR glasses and/or an AR input device may swipe through the representations in a stereoscopic manner.--, in [0024]; and, --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033]). Re Claim 16, KARANAM as modified by LIU further disclose wherein the scanning assembly comprises: a positioning assembly, positioning the scan object according to the keypoint distribution information (see KARANAM: e.g., Fig. 1, and --[0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: --scan images previously taken for the patient, positions of the patient during those scans, adjustments or corrections made to get the patient into a desired scan position, overlaid scan images and pictures (or models) of the patient, etc. The personalized medical assistance information may be displayed in various formats including, for example, videos, animations, and/or AR presentations. For example, the overlaid representations of the patient's scan images and pictures may be displayed in an AR environment in which a physician equipped with AR glasses and/or an AR input device may swipe through the representations in a stereoscopic manner.--, in [0024]; and, --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033]) . . 07-21-aia AIA Claim s 6-12 are rejected under 35 U.S.C. 103 as being unpatentable over KARANAM as modified by LIU further disclose, and further in view of Xie (US 20210074003 A1) . Re Claim 6, KARANAM as modified by LIU further disclose the keypoint image sequence comprises a plurality of keypoint images, wherein each of the keypoint images comprises a color keypoint image and a depth data corresponding to each other (see LIU: e.g., -- each data pair formed by a RGB and a RGBD is an image frame corresponding to the same view. Body keypoints of each frame of image in the RGB image data stream are aligned to depth data of body keypoints corresponding to the same image, so that for any one body keypoint in the image, both a 2D coordinate for characterizing a position of the body keypoint and a distance value for characterizing a change of movement of the body keypoint are obtained, and thus 3D information for the body keypoint position is obtained. The depth data are acquired from a DepthMap. In an acquisition scene including a camera or a camera module, the DepthMap may be considered as: an image (or referred as an image channel) that includes information on the distance to a surface of a target object acquired in the scene. When the distance of at least one point relative to the camera or the camera module in the scene is represented by a DepthMap, each pixel value in the DepthMap may represent a distance between a point and the camera in the scene.--, in line 57, col. 4, through line 8, col. 5; ; also see: -- the 3D pose data may be obtained by fusing the 2D pose data and depth data. The 2D pose data are 2D coordinates of body keypoints in an RGB image, and the 3D pose data are 3D body keypoints. By means of the 3D pose data, detection of keypoints of a human body may be performed accurately when the human body is in a motion state. For example, a motion state is decomposed into at least one node pose of raising a hand, kicking a leg, shaking a head or bending down, so that body keypoint detection corresponding to these node poses is tracked in real time.--, in lines 15-25, col. 4; also see: -- . For the first type of data (RGB data), after the first image processing, whether a human body is detected in the present image frame is determined by using a trained body tracking network; and if the body is detected, corresponding target RGB data in the present image frame are processed in a subsequent step. In the subsequent step, for the second type of data (RGBD data), after the second image processing, the RGBD data are fused with the target RGB data to obtain 3D pose data (3D coordinates of body skeleton keypoints) according to the RGBD data and the target RGB data. Dynamic tracking includes: using the 3D coordinates to represent the 3D pose data of the body skeleton keypoints to implement tracking when the body is in the motion state, such as tracking a change of node pose pairs, which supports at least one body action of raising the hand, kicking the leg, shaking the head, bending down, etc….. (32) FIG. 3 illustrates a schematic diagram of body skeleton keypoints according to an embodiment of the present disclosure. The body skeleton includes the number of 17 keypoints; and by detection of the 17 keypoints, a dynamic pose change of the user, such as at least one body action of raising the hand, kicking the leg, shaking the head, bending down and the like, may be tracked in real time. 33) For instance, in a scenario on how the user holding the mobile phone terminal interacts with a large-screen device such as a television, a first body motion state (such as a swing action when the user plays tennis) is acquired; and a change of body keypoint positions corresponding to the first body motion state is described through first 3D pose data. --, in line 30, col 5 through line 17, col. 6); KARANAM as modified by LIU however still do not explicitly disclose each of the keypoint images comprises a color keypoint image and a depth keypoint image corresponding to each other; Xie discloses each of the keypoint images comprises a color keypoint image and a depth keypoint image corresponding to each other (see Xie: e.g., -- the 3D image obtained in the step S110 includes: a 2D image and a depth image corresponding thereto, where the 2D image can provide a coordinate value of the skeleton keypoint in the xoy plane, and the depth value in the depth image can provide the coordinate of the skeleton keypoint on a z-axis. The z-axis is perpendicular to the xoy plane. [0128] A 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image is adjusted based on the second 3D coordinate so as to obtain the third 3D coordinate. [0129] Furthermore, obtaining the third 3D coordinate based on the second 3D coordinate includes: adjusting, based on the second 3D coordinate, a 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image so as to obtain the third 3D coordinate .--, in [0127]-[0131]); KARANAM (as modified by LIU) and Xie are combinable as they are in the same field of endeavor: monitoring and detection of keypoints of a patient/human body. Therefore it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to further modify KARANAM (as modified by LIU)’s method using Xie’s teachings by including each of the keypoint images comprises a color keypoint image and a depth keypoint image corresponding to each other to KARANAM (as modified by LIU)’s image processing and keypoints detection to associate the color keypoint image with the depth keypoint image such that the depth value in the depth image can provide the coordinate of the skeleton keypoint on a z-axis (see Xie: e.g., in [0127]-[0131]); Re Claim 7, KARANAM as modified by LIU and Xie further disclose the keypoint occlusion detection comprises detecting an occlusion state of a keypoint and time information of a change in the occlusion state (see KARANAM: e.g., --0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033];also see Xie: e.g., -- the 3D image obtained in the step S110 includes: a 2D image and a depth image corresponding thereto, where the 2D image can provide a coordinate value of the skeleton keypoint in the xoy plane, and the depth value in the depth image can provide the coordinate of the skeleton keypoint on a z-axis. The z-axis is perpendicular to the xoy plane. [0128] A 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image is adjusted based on the second 3D coordinate so as to obtain the third 3D coordinate. [0129] Furthermore, obtaining the third 3D coordinate based on the second 3D coordinate includes: adjusting, based on the second 3D coordinate, a 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image so as to obtain the third 3D coordinate . [0130] For example, a user laterally faces an image acquisition module; the depth value of the position of knees of both legs in the depth image is the same. In this case, the knee closer to the image acquisition module shields the knee relatively distant from the image acquisition module. In order to reduce the problem of inaccurate extraction of the 3D coordinate in the depth image caused by the shielding, the 3D coordinate for more precisely representing first-type movement information can be obtained by adjusting the 3D coordinate of a skeleton keypoint by using a deep learning model, a machine learning model, etc.--, in [0127]-[0131]). See the similar obviousness and motivation statements for the combination of cited references as addressed above for claim 6. Re Claim 8, KARANAM as modified by LIU and Xie further disclose wherein determining the keypoint distribution information according to the keypoint image sequence and the result of the keypoint occlusion detection comprises: determining a keypoint-occluded image from the keypoint image sequence; according to the occlusion state of the keypoint, determining, from the keypoint-occluded image, a first keypoint that is occluded and a second keypoint that is unoccluded (see KARANAM: e.g., --0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033];also see Xie: e.g., -- the 3D image obtained in the step S110 includes: a 2D image and a depth image corresponding thereto, where the 2D image can provide a coordinate value of the skeleton keypoint in the xoy plane, and the depth value in the depth image can provide the coordinate of the skeleton keypoint on a z-axis. The z-axis is perpendicular to the xoy plane. [0128] A 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image is adjusted based on the second 3D coordinate so as to obtain the third 3D coordinate. [0129] Furthermore, obtaining the third 3D coordinate based on the second 3D coordinate includes: adjusting, based on the second 3D coordinate, a 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image so as to obtain the third 3D coordinate . [0130] For example, a user laterally faces an image acquisition module; the depth value of the position of knees of both legs in the depth image is the same. In this case, the knee closer to the image acquisition module shields the knee relatively distant from the image acquisition module. In order to reduce the problem of inaccurate extraction of the 3D coordinate in the depth image caused by the shielding, the 3D coordinate for more precisely representing first-type movement information can be obtained by adjusting the 3D coordinate of a skeleton keypoint by using a deep learning model, a machine learning model, etc.--, in [0127]-[0131]); determining, according to time information of a change in an occlusion state of the first keypoint, a keypoint-unoccluded image from the keypoint image sequence (see KARANAM: e.g., --[0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033];also see Xie: e.g., -- the 3D image obtained in the step S110 includes: a 2D image and a depth image corresponding thereto, where the 2D image can provide a coordinate value of the skeleton keypoint in the xoy plane, and the depth value in the depth image can provide the coordinate of the skeleton keypoint on a z-axis. The z-axis is perpendicular to the xoy plane. [0128] A 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image is adjusted based on the second 3D coordinate so as to obtain the third 3D coordinate. [0129] Furthermore, obtaining the third 3D coordinate based on the second 3D coordinate includes: adjusting, based on the second 3D coordinate, a 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image so as to obtain the third 3D coordinate . [0130] For example, a user laterally faces an image acquisition module; the depth value of the position of knees of both legs in the depth image is the same. In this case, the knee closer to the image acquisition module shields the knee relatively distant from the image acquisition module. In order to reduce the problem of inaccurate extraction of the 3D coordinate in the depth image caused by the shielding, the 3D coordinate for more precisely representing first-type movement information can be obtained by adjusting the 3D coordinate of a skeleton keypoint by using a deep learning model, a machine learning model, etc.--, in [0127]-[0131]); and determining the keypoint distribution information according to the keypoint-unoccluded image and the keypoint-occluded image (see Xie: e.g., --position information of keypoints of the body skeleton of the whole body can be first obtained by using deep learning models such as a neural network so as to determine position information of a keypoint of the limb based on the distribution of the position information of the keypoints of the whole body. After the position information of the keypoints of the whole body is obtained, the keypoints are connected to obtain the skeleton, and it can be determined which keypoints are keypoints of the limb based on the relative distribution position of bones and joints in the skeleton so as to determine the position information of the keypoint of the limb.--, in [0062]). Re Claim 9, KARANAM as modified by LIU and Xie further disclose the keypoint-occluded image is the last keypoint image in the keypoint image sequence (see KARANAM: e.g., --[0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033]; and see LIU: e.g., -- the 3D pose data may be obtained by fusing the 2D pose data and depth data. The 2D pose data are 2D coordinates of body keypoints in an RGB image, and the 3D pose data are 3D body keypoints. By means of the 3D pose data, detection of keypoints of a human body may be performed accurately when the human body is in a motion state. For example, a motion state is decomposed into at least one node pose of raising a hand, kicking a leg, shaking a head or bending down, so that body keypoint detection corresponding to these node poses is tracked in real time.--, in lines 15-25, col. 4; also see: -- . For the first type of data (RGB data), after the first image processing, whether a human body is detected in the present image frame is determined by using a trained body tracking network; and if the body is detected, corresponding target RGB data in the present image frame are processed in a subsequent step. In the subsequent step, for the second type of data (RGBD data), after the second image processing, the RGBD data are fused with the target RGB data to obtain 3D pose data (3D coordinates of body skeleton keypoints) according to the RGBD data and the target RGB data. Dynamic tracking includes: using the 3D coordinates to represent the 3D pose data of the body skeleton keypoints to implement tracking when the body is in the motion state, such as tracking a change of node pose pairs, which supports at least one body action of raising the hand, kicking the leg, shaking the head, bending down, etc ….. (32) FIG. 3 illustrates a schematic diagram of body skeleton keypoints according to an embodiment of the present disclosure. The body skeleton includes the number of 17 keypoints; and by detection of the 17 keypoints, a dynamic pose change of the user , such as at least one body action of raising the hand, kicking the leg, shaking the head, bending down and the like, may be tracked in real time. 33) For instance, in a scenario on how the user holding the mobile phone terminal interacts with a large-screen device such as a television, a first body motion state (such as a swing action when the user plays tennis) is acquired; and a change of body keypoint positions corresponding to the first body motion state is described through first 3D pose data. --, in line 30, col 5 through line 17, col. 6; also see Xie: e.g., -- the 3D image obtained in the step S110 includes: a 2D image and a depth image corresponding thereto, where the 2D image can provide a coordinate value of the skeleton keypoint in the xoy plane, and the depth value in the depth image can provide the coordinate of the skeleton keypoint on a z-axis. The z-axis is perpendicular to the xoy plane. [0128] A 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image is adjusted based on the second 3D coordinate so as to obtain the third 3D coordinate. [0129] Furthermore, obtaining the third 3D coordinate based on the second 3D coordinate includes: adjusting, based on the second 3D coordinate, a 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image so as to obtain the third 3D coordinate . [0130] For example, a user laterally faces an image acquisition module; the depth value of the position of knees of both legs in the depth image is the same. In this case, the knee closer to the image acquisition module shields the knee relatively distant from the image acquisition module. In order to reduce the problem of inaccurate extraction of the 3D coordinate in the depth image caused by the shielding, the 3D coordinate for more precisely representing first-type movement information can be obtained by adjusting the 3D coordinate of a skeleton keypoint by using a deep learning model, a machine learning model, etc.--, in [0127]-[0131]). Re Claim 10, KARANAM as modified by LIU and Xie further disclose in the keypoint-unoccluded image, keypoints of the same type as the first keypoint are unoccluded, and in the time dimension of the keypoint image sequence, the keypoint-unoccluded image is closest to the keypoint-occluded image (see KARANAM: e.g., --[0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033]; and see LIU: e.g., -- the 3D pose data may be obtained by fusing the 2D pose data and depth data. The 2D pose data are 2D coordinates of body keypoints in an RGB image, and the 3D pose data are 3D body keypoints. By means of the 3D pose data, detection of keypoints of a human body may be performed accurately when the human body is in a motion state. For example, a motion state is decomposed into at least one node pose of raising a hand, kicking a leg, shaking a head or bending down, so that body keypoint detection corresponding to these node poses is tracked in real time.--, in lines 15-25, col. 4; also see: -- . For the first type of data (RGB data), after the first image processing, whether a human body is detected in the present image frame is determined by using a trained body tracking network; and if the body is detected, corresponding target RGB data in the present image frame are processed in a subsequent step. In the subsequent step, for the second type of data (RGBD data), after the second image processing, the RGBD data are fused with the target RGB data to obtain 3D pose data (3D coordinates of body skeleton keypoints) according to the RGBD data and the target RGB data. Dynamic tracking includes: using the 3D coordinates to represent the 3D pose data of the body skeleton keypoints to implement tracking when the body is in the motion state, such as tracking a change of node pose pairs, which supports at least one body action of raising the hand, kicking the leg, shaking the head, bending down, etc ….. (32) FIG. 3 illustrates a schematic diagram of body skeleton keypoints according to an embodiment of the present disclosure. The body skeleton includes the number of 17 keypoints; and by detection of the 17 keypoints, a dynamic pose change of the user , such as at least one body action of raising the hand, kicking the leg, shaking the head, bending down and the like, may be tracked in real time. 33) For instance, in a scenario on how the user holding the mobile phone terminal interacts with a large-screen device such as a television, a first body motion state (such as a swing action when the user plays tennis) is acquired; and a change of body keypoint positions corresponding to the first body motion state is described through first 3D pose data. --, in line 30, col 5 through line 17, col. 6; also see Xie: e.g., -- the 3D image obtained in the step S110 includes: a 2D image and a depth image corresponding thereto, where the 2D image can provide a coordinate value of the skeleton keypoint in the xoy plane, and the depth value in the depth image can provide the coordinate of the skeleton keypoint on a z-axis. The z-axis is perpendicular to the xoy plane. [0128] A 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image is adjusted based on the second 3D coordinate so as to obtain the third 3D coordinate. [0129] Furthermore, obtaining the third 3D coordinate based on the second 3D coordinate includes: adjusting, based on the second 3D coordinate, a 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image so as to obtain the third 3D coordinate . [0130] For example, a user laterally faces an image acquisition module; the depth value of the position of knees of both legs in the depth image is the same. In this case, the knee closer to the image acquisition module shields the knee relatively distant from the image acquisition module. In order to reduce the problem of inaccurate extraction of the 3D coordinate in the depth image caused by the shielding, the 3D coordinate for more precisely representing first-type movement information can be obtained by adjusting the 3D coordinate of a skeleton keypoint by using a deep learning model, a machine learning model, etc.--, in [0127]-[0131]). Re Claim 11, KARANAM as modified by LIU and Xie further disclose wherein determining the keypoint distribution information according to the keypoint-unoccluded image and the keypoint-occluded image comprises: determining, from the keypoint-unoccluded image, a third keypoint that is unoccluded, the third keypoint being of the same type as the first keypoint that is occluded (see KARANAM: e.g., --[0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033]; and see LIU: e.g., -- the 3D pose data may be obtained by fusing the 2D pose data and depth data. The 2D pose data are 2D coordinates of body keypoints in an RGB image, and the 3D pose data are 3D body keypoints. By means of the 3D pose data, detection of keypoints of a human body may be performed accurately when the human body is in a motion state. For example, a motion state is decomposed into at least one node pose of raising a hand, kicking a leg, shaking a head or bending down, so that body keypoint detection corresponding to these node poses is tracked in real time.--, in lines 15-25, col. 4; also see: -- . For the first type of data (RGB data), after the first image processing, whether a human body is detected in the present image frame is determined by using a trained body tracking network; and if the body is detected, corresponding target RGB data in the present image frame are processed in a subsequent step. In the subsequent step, for the second type of data (RGBD data), after the second image processing, the RGBD data are fused with the target RGB data to obtain 3D pose data (3D coordinates of body skeleton keypoints) according to the RGBD data and the target RGB data. Dynamic tracking includes: using the 3D coordinates to represent the 3D pose data of the body skeleton keypoints to implement tracking when the body is in the motion state, such as tracking a change of node pose pairs, which supports at least one body action of raising the hand, kicking the leg, shaking the head, bending down, etc ….. (32) FIG. 3 illustrates a schematic diagram of body skeleton keypoints according to an embodiment of the present disclosure. The body skeleton includes the number of 17 keypoints; and by detection of the 17 keypoints, a dynamic pose change of the user , such as at least one body action of raising the hand, kicking the leg, shaking the head, bending down and the like, may be tracked in real time. 33) For instance, in a scenario on how the user holding the mobile phone terminal interacts with a large-screen device such as a television, a first body motion state (such as a swing action when the user plays tennis) is acquired; and a change of body keypoint positions corresponding to the first body motion state is described through first 3D pose data. --, in line 30, col 5 through line 17, col. 6; also see Xie: e.g., -- the 3D image obtained in the step S110 includes: a 2D image and a depth image corresponding thereto, where the 2D image can provide a coordinate value of the skeleton keypoint in the xoy plane, and the depth value in the depth image can provide the coordinate of the skeleton keypoint on a z-axis. The z-axis is perpendicular to the xoy plane. [0128] A 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image is adjusted based on the second 3D coordinate so as to obtain the third 3D coordinate. [0129] Furthermore, obtaining the third 3D coordinate based on the second 3D coordinate includes: adjusting, based on the second 3D coordinate, a 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image so as to obtain the third 3D coordinate . [0130] For example, a user laterally faces an image acquisition module; the depth value of the position of knees of both legs in the depth image is the same. In this case, the knee closer to the image acquisition module shields the knee relatively distant from the image acquisition module. In order to reduce the problem of inaccurate extraction of the 3D coordinate in the depth image caused by the shielding, the 3D coordinate for more precisely representing first-type movement information can be obtained by adjusting the 3D coordinate of a skeleton keypoint by using a deep learning model, a machine learning model, etc.--, in [0127]-[0131]); and generating the keypoint distribution information according to the third keypoint in the keypoint-unoccluded image and the second keypoint that is unoccluded in the keypoint-occluded image (see Xie: e.g., --position information of keypoints of the body skeleton of the whole body can be first obtained by using deep learning models such as a neural network so as to determine position information of a keypoint of the limb based on the distribution of the position information of the keypoints of the whole body. After the position information of the keypoints of the whole body is obtained, the keypoints are connected to obtain the skeleton, and it can be determined which keypoints are keypoints of the limb based on the relative distribution position of bones and joints in the skeleton so as to determine the position information of the keypoint of the limb.--, in [0062]). Re Claim 12, KARANAM as modified by LIU and Xie further disclose wherein generating the keypoint distribution information according to the third keypoint in the keypoint-unoccluded image and the second keypoint that is unoccluded in the keypoint-occluded image (see KARANAM: e.g., --[0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033]; and see LIU: e.g., -- the 3D pose data may be obtained by fusing the 2D pose data and depth data. The 2D pose data are 2D coordinates of body keypoints in an RGB image, and the 3D pose data are 3D body keypoints. By means of the 3D pose data, detection of keypoints of a human body may be performed accurately when the human body is in a motion state. For example, a motion state is decomposed into at least one node pose of raising a hand, kicking a leg, shaking a head or bending down, so that body keypoint detection corresponding to these node poses is tracked in real time.--, in lines 15-25, col. 4; also see: -- . For the first type of data (RGB data), after the first image processing, whether a human body is detected in the present image frame is determined by using a trained body tracking network; and if the body is detected, corresponding target RGB data in the present image frame are processed in a subsequent step. In the subsequent step, for the second type of data (RGBD data), after the second image processing, the RGBD data are fused with the target RGB data to obtain 3D pose data (3D coordinates of body skeleton keypoints) according to the RGBD data and the target RGB data. Dynamic tracking includes: using the 3D coordinates to represent the 3D pose data of the body skeleton keypoints to implement tracking when the body is in the motion state, such as tracking a change of node pose pairs, which supports at least one body action of raising the hand, kicking the leg, shaking the head, bending down, etc ….. (32) FIG. 3 illustrates a schematic diagram of body skeleton keypoints according to an embodiment of the present disclosure. The body skeleton includes the number of 17 keypoints; and by detection of the 17 keypoints, a dynamic pose change of the user , such as at least one body action of raising the hand, kicking the leg, shaking the head, bending down and the like, may be tracked in real time. 33) For instance, in a scenario on how the user holding the mobile phone terminal interacts with a large-screen device such as a television, a first body motion state (such as a swing action when the user plays tennis) is acquired; and a change of body keypoint positions corresponding to the first body motion state is described through first 3D pose data. --, in line 30, col 5 through line 17, col. 6; also see Xie: e.g., -- the 3D image obtained in the step S110 includes: a 2D image and a depth image corresponding thereto, where the 2D image can provide a coordinate value of the skeleton keypoint in the xoy plane, and the depth value in the depth image can provide the coordinate of the skeleton keypoint on a z-axis. The z-axis is perpendicular to the xoy plane. [0128] A 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image is adjusted based on the second 3D coordinate so as to obtain the third 3D coordinate. [0129] Furthermore, obtaining the third 3D coordinate based on the second 3D coordinate includes: adjusting, based on the second 3D coordinate, a 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image so as to obtain the third 3D coordinate . [0130] For example, a user laterally faces an image acquisition module; the depth value of the position of knees of both legs in the depth image is the same. In this case, the knee closer to the image acquisition module shields the knee relatively distant from the image acquisition module. In order to reduce the problem of inaccurate extraction of the 3D coordinate in the depth image caused by the shielding, the 3D coordinate for more precisely representing first-type movement information can be obtained by adjusting the 3D coordinate of a skeleton keypoint by using a deep learning model, a machine learning model, etc.--, in [0127]-[0131]) comprises: updating position information of the first keypoint in the keypoint-occluded image by using position information of the third keypoint in the keypoint-unoccluded image (see KARANAM: e.g., --[0028] The features and/or characteristics described herein may be associated with a variety of attributes of the patient such as body contour, height, facial features, walking patterns, poses, etc. In the context of digital imagery, these features or characteristics may correspond to structures in an image such as points, edges, objects, etc. Various techniques may be employed to extract these features from the image. For example, one or more keypoints associated with a feature may be identified including points at which the direction of the boundary of an object changes abruptly, intersection points between two or more edge segments, etc. These keypoints may be characterized by well-defined positions in the image space and/or stability to illumination/brightness perturbations. As such, the keypoints may be identified based on image derivatives, edge detection, curvature analysis, and/or the like. [0029] Once identified, the keypoints and/or the feature associated with the keypoints may be described with a feature descriptor or feature vector. In an example implementation of such feature descriptor or vector, information related to the feature (e.g., appearance of the local neighborhood of each keypoint) may be represented by (e.g., encoded into) a series of numerical values stored in the feature descriptor or vector. The descriptor or vector may then be used as a “fingerprint” for differentiating one feature from another or matching one feature with another….. Further, the processor 202 may be configured to determine that certain areas of the patient's body are obstructed or occluded, and subsequently avoid using features associated with the obstructed areas for patient matching (e.g., the processor 202 may decide to use a different feature such as the walking pattern of the patient for identifying the patient). Obstructed or occluded areas of the patient may be determined, for example, by running occlusion detectors for one or more parts of the patient body (e.g., in a bottom-up manner), and/or by recognizing the overall pose of the patient and then inferring the occluded areas based on the overall pose of the patient. Depth information associated with one or more images of the patient may be used to determine the obstructed or occluded areas.--, in [0028]-[0030]; also see: --once a matching patient is found, the processor 202 may proceed to query a repository (e.g., the repository 112 in FIG. 1) to retrieve medical records (e.g., imagery and/or non-imagery data associated with a medical procedure) of the patient based on the patient's identity. The medical records may include, for example, positioning information associated with a medical procedure to be performed for the patient, previous scan pictures or other types of images of the patient, diagnostic and treatment history of the patient, etc. The processor 202 may generate personalized medical assistance information (e.g., build a medical profile) for the patient based on the retrieved medical records.--, in [0032]-[0033]; and see LIU: e.g., -- the 3D pose data may be obtained by fusing the 2D pose data and depth data. The 2D pose data are 2D coordinates of body keypoints in an RGB image, and the 3D pose data are 3D body keypoints. By means of the 3D pose data, detection of keypoints of a human body may be performed accurately when the human body is in a motion state. For example, a motion state is decomposed into at least one node pose of raising a hand, kicking a leg, shaking a head or bending down, so that body keypoint detection corresponding to these node poses is tracked in real time.--, in lines 15-25, col. 4; also see: -- . For the first type of data (RGB data), after the first image processing, whether a human body is detected in the present image frame is determined by using a trained body tracking network; and if the body is detected, corresponding target RGB data in the present image frame are processed in a subsequent step. In the subsequent step, for the second type of data (RGBD data), after the second image processing, the RGBD data are fused with the target RGB data to obtain 3D pose data (3D coordinates of body skeleton keypoints) according to the RGBD data and the target RGB data. Dynamic tracking includes: using the 3D coordinates to represent the 3D pose data of the body skeleton keypoints to implement tracking when the body is in the motion state, such as tracking a change of node pose pairs, which supports at least one body action of raising the hand, kicking the leg, shaking the head, bending down, etc ….. (32) FIG. 3 illustrates a schematic diagram of body skeleton keypoints according to an embodiment of the present disclosure. The body skeleton includes the number of 17 keypoints; and by detection of the 17 keypoints, a dynamic pose change of the user , such as at least one body action of raising the hand, kicking the leg, shaking the head, bending down and the like, may be tracked in real time. 33) For instance, in a scenario on how the user holding the mobile phone terminal interacts with a large-screen device such as a television, a first body motion state (such as a swing action when the user plays tennis) is acquired; and a change of body keypoint positions corresponding to the first body motion state is described through first 3D pose data. --, in line 30, col 5 through line 17, col. 6; also see Xie: e.g., -- the 3D image obtained in the step S110 includes: a 2D image and a depth image corresponding thereto, where the 2D image can provide a coordinate value of the skeleton keypoint in the xoy plane, and the depth value in the depth image can provide the coordinate of the skeleton keypoint on a z-axis. The z-axis is perpendicular to the xoy plane. [0128] A 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image is adjusted based on the second 3D coordinate so as to obtain the third 3D coordinate. [0129] Furthermore, obtaining the third 3D coordinate based on the second 3D coordinate includes: adjusting, based on the second 3D coordinate, a 3D coordinate of the skeleton keypoint of the limb corresponding to the occluded portion in the 3D image so as to obtain the third 3D coordinate . [0130] For example, a user laterally faces an image acquisition module; the depth value of the position of knees of both legs in the depth image is the same. In this case, the knee closer to the image acquisition module shields the knee relatively distant from the image acquisition module. In order to reduce the problem of inaccurate extraction of the 3D coordinate in the depth image caused by the shielding, the 3D coordinate for more precisely representing first-type movement information can be obtained by adjusting the 3D coordinate of a skeleton keypoint by using a deep learning model, a machine learning model, etc.--, in [0127]-[0131]); and generating the keypoint distribution information according to the updated position information of the first keypoint and position information of the second keypoint (see Xie: e.g., --position information of keypoints of the body skeleton of the whole body can be first obtained by using deep learning models such as a neural network so as to determine position information of a keypoint of the limb based on the distribution of the position information of the keypoints of the whole body. After the position information of the keypoints of the whole body is obtained, the keypoints are connected to obtain the skeleton, and it can be determined which keypoints are keypoints of the limb based on the relative distribution position of bones and joints in the skeleton so as to determine the position information of the keypoint of the limb.--, in [0062]). Conclusion Any inquiry concerning this communication or earlier communications from the examiner should be directed to WEIWEN YANG whose telephone number is (571)270-5670. The examiner can normally be reached on Monday-Friday 8:30am-4:30pm east. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Amandeep Saini can be reached on 571-272-3382. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300. Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000. /WEI WEN YANG/Primary Examiner, Art Unit 2662 Application/Control Number: 18/883,814 Page 2 Art Unit: 2667 Application/Control Number: 18/883,814 Page 3 Art Unit: 2667 Application/Control Number: 18/883,814 Page 4 Art Unit: 2667 Application/Control Number: 18/883,814 Page 5 Art Unit: 2667 Application/Control Number: 18/883,814 Page 6 Art Unit: 2667 Application/Control Number: 18/883,814 Page 7 Art Unit: 2667 Application/Control Number: 18/883,814 Page 8 Art Unit: 2667 Application/Control Number: 18/883,814 Page 9 Art Unit: 2667 Application/Control Number: 18/883,814 Page 10 Art Unit: 2667 Application/Control Number: 18/883,814 Page 11 Art Unit: 2667 Application/Control Number: 18/883,814 Page 12 Art Unit: 2667 Application/Control Number: 18/883,814 Page 13 Art Unit: 2667 Application/Control Number: 18/883,814 Page 14 Art Unit: 2667 Application/Control Number: 18/883,814 Page 15 Art Unit: 2667 Application/Control Number: 18/883,814 Page 16 Art Unit: 2667 Application/Control Number: 18/883,814 Page 17 Art Unit: 2667 Application/Control Number: 18/883,814 Page 18 Art Unit: 2667 Application/Control Number: 18/883,814 Page 19 Art Unit: 2667 Application/Control Number: 18/883,814 Page 20 Art Unit: 2667 Application/Control Number: 18/883,814 Page 21 Art Unit: 2667 Application/Control Number: 18/883,814 Page 22 Art Unit: 2667 Application/Control Number: 18/883,814 Page 23 Art Unit: 2667 Application/Control Number: 18/883,814 Page 24 Art Unit: 2667 Application/Control Number: 18/883,814 Page 25 Art Unit: 2667 Application/Control Number: 18/883,814 Page 26 Art Unit: 2667 Application/Control Number: 18/883,814 Page 27 Art Unit: 2667 Application/Control Number: 18/883,814 Page 28 Art Unit: 2667 Application/Control Number: 18/883,814 Page 29 Art Unit: 2667 Application/Control Number: 18/883,814 Page 30 Art Unit: 2667 Application/Control Number: 18/883,814 Page 31 Art Unit: 2667 Application/Control Number: 18/883,814 Page 32 Art Unit: 2667 Application/Control Number: 18/883,814 Page 33 Art Unit: 2667 Application/Control Number: 18/883,814 Page 34 Art Unit: 2667 Application/Control Number: 18/883,814 Page 35 Art Unit: 2667 Application/Control Number: 18/883,814 Page 36 Art Unit: 2667 Application/Control Number: 18/883,814 Page 37 Art Unit: 2667 Application/Control Number: 18/883,814 Page 38 Art Unit: 2667 Application/Control Number: 18/883,814 Page 39 Art Unit: 2667 Application/Control Number: 18/883,814 Page 40 Art Unit: 2667 Application/Control Number: 18/883,814 Page 41 Art Unit: 2667 Application/Control Number: 18/883,814 Page 42 Art Unit: 2667 Application/Control Number: 18/883,814 Page 43 Art Unit: 2667 Application/Control Number: 18/883,814 Page 44 Art Unit: 2667 Application/Control Number: 18/883,814 Page 45 Art Unit: 2667 Application/Control Number: 18/883,814 Page 46 Art Unit: 2667 Application/Control Number: 18/883,814 Page 47 Art Unit: 2667 Application/Control Number: 18/883,814 Page 48 Art Unit: 2667 Application/Control Number: 18/883,814 Page 49 Art Unit: 2667 Application/Control Number: 18/883,814 Page 50 Art Unit: 2667 Application/Control Number: 18/883,814 Page 51 Art Unit: 2667 Application/Control Number: 18/883,814 Page 52 Art Unit: 2667 Application/Control Number: 18/883,814 Page 53 Art Unit: 2667 Application/Control Number: 18/883,814 Page 54 Art Unit: 2667 Application/Control Number: 18/883,814 Page 55 Art Unit: 2667 Application/Control Number: 18/883,814 Page 56 Art Unit: 2667 Application/Control Number: 18/883,814 Page 57 Art Unit: 2667 Application/Control Number: 18/883,814 Page 58 Art Unit: 2667
Prosecution Insights