DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55 (CN202010327417.4, CN202010374378.3).
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's RCE submission filed on 01/08/2026 has been entered.
Response to Amendment
The amendment filed 01/08/2026 has been entered. Applicant’s amendments to claims 1, 6-7, 13, 21, 24-25, 34 and 36 have overcome each and every claim objection previously set forth in the Final Office Action mailed 10/08/2025. The claim objection for claim 5 remains outstanding below. Applicant’s amendments to the claims have overcome each and every 35 U.S.C. 112(b) rejection previously set forth in the Final Office Action mailed 10/08/2025. Claims 1-8, 13-14, 21, 24-26, 28 and 34-38 remain pending in the application, with claims 16 and 33 having been cancelled and claims 37-38 being new.
Response to Arguments
Pg. 14 of the Remarks, filed 01/08/2026, addresses claims allegedly interpreted under 35 U.S.C. 112(f). There are no claims being interpreted under 35 U.S.C. 112(f).
Regarding claims 1-4 and 6-7, Applicant's arguments filed 01/08/2026 have been fully considered but they are not persuasive. In the last paragraph of pg. 15, Applicant argues: “Applicant respectfully submits that Guehring, Abdolell, and Choi, either alone or in combination, do not disclose each and every element set forth in amended claim 1.” Examiner respectfully disagrees. Although amended claim 1 lists five different evaluation models, each used to obtain a different score related to types of evaluations indexed, the five different evaluation models are not required by the claim because claim 1 recites “one or more evaluation indexes” (emphasis added). For example, in a situation wherein two evaluation indexes are utilized to generate scores, at least two different evaluation models (required by “multiple”) are required, but all five recited evaluation models are not. In a similar case, if the evaluation index “a window width” is utilized, none of the five recited evaluation models apply to obtain the relevant score. Thus, even if Guehring, Abdolell, and Choi, either alone or in combination, do not disclose the five recited evaluation models listed in the remarks on pg. 15, they may still teach the invention of claim 1.
Furthermore, the Examiner submits that the combination of Guehring in view of Abdolell and Choi to reject claim 1 is maintained. On pg. 16-17 of the remarks, Applicant argues that Choi does not disclose “the multiple different evaluation models are trained based on different training samples” in claim 1 in view of the underlined arguments attached below.
From pg. 16:
PNG
media_image1.png
151
736
media_image1.png
Greyscale
From pg. 17:
PNG
media_image2.png
154
732
media_image2.png
Greyscale
Examiner respectfully disagrees. Choi teaches two neural networks trained on different training data sets, thus teaching wherein multiple different evaluation models are trained based on different training samples. Choi is solely relied upon to teach this limitation, which is simply a characteristic of how multiple models are trained, in combination with relied upon prior art references. Because Guehring in view of Abdolell is relied upon to teach the different evaluation models that evaluate the image quality indexes, it is irrelevant that Choi’s models solve a different problem, since the technical solution is addressed in the combination. This reasoning is similarly applied to the arguments in the last paragraph on pg. 17. In view of the combination, the model training is considered as a whole. One cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986).
On pg. 18-19 of the remarks, Applicant argues that Abdollel fails to disclose aspects of claim 1, in view of the following underlined statements:
PNG
media_image3.png
572
741
media_image3.png
Greyscale
Examiner respectfully disagrees. Abdolell is solely relied upon to teach wherein multiple different evaluation models output multiple scores related to evaluation indexes. This is in combination with relied upon prior art reference Guehring, specifically including the evaluation indexes taught by Guehring. Evaluating the evaluation indexes taught by Guehring using multiple different predictive models improves the prediction accuracy, as cited in the rejection. Because Guehring in view of Abdolell is relied upon to teach the relevant limitations of claim 1, the specific quality parameter features of Abdolell are irrelevant, since the technical solution is addressed in the combination. One cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986). Therefore, the rejection of Guehring in view of Abdolell and Choi in claims 1-4 and 6-7 is maintained below.
Regarding claim 5, for the reasons described above with respect to claim 1, the rejection of claim 5 is maintained.
Regarding claim 8, for the reasons described above with respect to claim 1, the rejection of claim 8 is maintained.
Regarding claim 13, for the reasons described above, Guehring in view of Abdollel and Choi teach the limitations of amended claim 1. The arguments on pg. 24-25 of the remarks are moot, because claim 13 relies on the newly introduced prior art reference, Fu, introduced in the rejection below.
Regarding claims 14 and 25, for the reasons described above with respect to claim 1, the rejection of claims 14 and 25 is maintained.
Regarding claim 21, the arguments with respect to claim 21, starting on pg. 27 through the beginning of pg. 31 are moot because claim 21 does not depend on claim 1 and is rejected using different prior art references (Shanghai United in view of Profio). The arguments are directed towards the combination with Guehring, among other references, which claim 21 does not rely on. Additionally, claim 24 does not appear to be considered cancelled, and remains pending (see top of pg. 28 of the remarks). Specific remarks regarding the teaching of Kitamura will be addressed with respect to claim 24 below.
Regarding claims 24 and 34, pg. 31-32 of the remarks, for the reasons described above, Guehring in view of Abdollel and Choi teach the limitations of amended claim 1. Regarding claim 34 individually, Applicant’s arguments on pg. 33-34 are moot because claim 34 relies on the newly introduced prior art reference, Fu, introduced in the rejection below.
Regarding claim 24 individually, Applicant's arguments, see pg. 28-31 and the last paragraph on pg. 32, with respect to claim 24 (specifically, the reference Kitamura) have been fully considered but are not persuasive. Applicant argues the following underlined portion on pg. 29:
PNG
media_image4.png
305
733
media_image4.png
Greyscale
Claim 24 fails to specify where the gantry is located relative to what is considered the imaging device, and thus doesn’t require the argued positional relationship. Even if Kitamura does not discuss how to align and associate its measured point cloud data with the coordinate system of another independent device, this itself is not recited as a required limitation in claim 24. The recitation of “the imaging device” can include the gantry of the imaging device as an entire system. Thus, if the scanner of Kitamura is incorporated in an imaging device, the relative distance and angle measurements between the subject and the imaging device reflects a relative position between the subject and the gantry. Details regarding the claimed relationship “is obtained based on” are not recited in either the claim or specification.
Additionally, Applicant argues that a person having ordinary skill in the art would not be motivated to combine Guehring/Profio with Kitamura. Examiner respectfully disagrees. Profio teaches generating a 3D “patient envelope” which conveys the 3D anatomy of the subject to be imaged. Furthermore, where the subject is positioned is an important factor in determining image quality and acquisition settings in the disclosure of Guehring. Kitamura determines a relative distance and angle between a scanner and an object. A person of ordinary skill in the art would recognize the advantage of having 3D positioning data of the subject to be imaged relative to a position in the imaging device.
Lastly, Applicant argues the following underlined portion on pg. 30:
PNG
media_image5.png
227
734
media_image5.png
Greyscale
The examiner submits that the use of laser technology to assist in the positioning of subjects in a medical imaging device is known in the art, as reflected by Booij, cited in the Final Office Action mailed 10/08/2025: Pg. 2028, “Patients were positioned on the scanner by the radiographer using laser beam guidance as routinely available on the CT scanner.”. In view of the foregoing, the examiner maintains the use of the relevant teachings of Kitamura in the combination rejecting claim 24 below. For similar reasons, the relevant teachings of Kitamura are combined with Shanghai United in view of Profio in the rejection of claim 21 below.
Regarding claim 26, for the reasons described above with respect to claim 1, the rejection of claim 26 is maintained.
Regarding claim 28, for the reasons described above with respect to claim 1, the rejection of claim 28 is maintained.
Regarding claims 35-36, pg. 37 of the remarks, for the reasons described above, Guehring in view of Abdollel and Choi teach the limitations of amended claim 1. The arguments on pg. 38-40 of the remarks are moot, because claims 35-36 rely on the newly introduced prior art reference, Fu, introduced in the rejection below.
Regarding claims 37-38, because the rejection of claim 1 is maintained due to reasons described above, claims 37-38 are addressed in the rejection below.
Claim Objections
Claims 5, 34, and 37-38 are objected to because of the following informalities:
Regarding claim 5, "wherein the posture information of the subject includes a relative position of an imaged portion" should read "wherein the posture information of the target subject includes a relative position of the imaged portion".
Claim 34 should depend on claim 13, not claim 1 (based on reference to the local features).
Regarding claim 37, “whose posture information has a difference with the posture information is within a threshold range, determining that there is the matching imaging protocol in the historical imaging protocols; or in response to there is no imaging protocol in the candidate imaging protocols that is the same as imaged portion and whose posture information has a difference with the posture information exceeds the threshold” should read “whose posture information has a difference with the posture information of the subject to be imaged that is within a threshold range, determining that there is a matching imaging protocol in the historical imaging protocols; or in response to there is no imaging protocol in the candidate imaging protocols that is the same as the imaged portion and whose posture information has a difference with the posture information of the subject to be imaged that exceeds the threshold”.
Regarding claim 38, “the one or more evaluation indexe” should read “the one or more evaluation indexes”.
Appropriate correction is required.
Claim Interpretation
Regarding claims 5 and 21, the posture information of the subject is interpreted to include at least one of all of the listed elements (a relative position of an imaged portion of the subject with respect to a gantry, a thickness of the imaged portion, a width of the imaged portion, and a height of the imaged portion determined based on three-dimensional contour data of the subject), in accordance with the Federal Circuit’s 2004 Superguide Corp. v. DirecTV Enterprises, Inc. decision.
Regarding claim 34, the one or more local features is interpreted to include at least one of all of the listed elements (color features, texture features, shape features, and local feature points), in accordance with the Federal Circuit’s 2004 Superguide Corp. v. DirecTV Enterprises, Inc. decision.
Regarding claim 37, “search keywords” are recited to include positioning information, which further includes “standing imaging, side lying image, and supine imaging”. Paragraph 160 of the specification states the following: “For example, the search keywords may be the positioning information of the subject to be imaged in the imaging protocol (for example, standing imaging, side lying imaging, supine imaging, etc.)”. Thus, “the search keywords” in claim 37 will be interpreted to be the following required terms: “standing imaging”, “side lying imaging”, and “supine imaging”.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claim 37 is rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 37 recites wherein “a difference with the posture information” is compared to a threshold range. As recited in independent claim 21, the posture information includes four different types of values (i.e., relative position, thickness, width, height). Thus, it is unclear in claim 37 which posture information values are used in the difference calculation and comparison to the threshold range. For examination purposes, claim 37 will be interpreted to use any one of the types of posture information in the difference calculation with the posture information of the candidate imaging protocols.
Additionally, claim 37 recites both “determining that there is the matching imaging protocol in the historical imaging protocols” and “determining that there is no matching imaging protocol in the historical imaging protocols” combined in the disjunctive form with the word “or”. Therefore, the plain language meaning of the claim only requires one of the recited claim limitations in claim 37, each separated by a semicolon. However, independent claim 21 has two “in response to” clauses directly based on the “determining” clauses in claim 37: “in response to determining that there is a matching imaging protocol in the matching imaging protocols” and “in response to determining that there are no matching imaging protocols in the historical imaging protocols”. Thus, it is unclear whether both determining clauses are required to be met in claim 37. In view of the foregoing, claim 37 will be interpreted to require all claim limitations, as if conjoined by “and” instead of “or”.
Claim Rejections - 35 USC § 103
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.
Claims 1-4 and 6-7 are rejected under 35 U.S.C. 103 as being unpatentable over Guehring et al. (U.S. Patent No. 2011/0110572 A1), hereinafter Guehring, in view of Abdolell et al. (U.S. Patent No. 2023/0071400 A1), hereinafter Abdolell, and Choi et al. (KR Patent No. 20200005413 A), hereinafter Choi.
Regarding claim 1, Guehring teaches a method for image acquisition (Guehring, para 14: “medical images acquired by medical imaging device”), implemented by a processing device (Guehring, para 4: “A system dynamically improves quality of medical images acquired by a medical imaging device using at least one processing device”), wherein the method comprises:
obtaining scanning data of a target subject by using an imaging device (Guehring, para 14: “medical images acquired by medical imaging device 40”);
obtaining an image of the target subject by processing the scanning data or from a storage device or a medical database (Guehring, para 14: “DICOM and other images 11 containing undesired image artifacts or exhibiting compromised image quality acquired by imaging device 40 are processed by an image processing unit 13”; see also Figure 1);
obtaining, by inputting the image of the target subject (Guehring, “analyzer 20” in para 18 and Fig. 1; para 16: “Image analyzer 20 automatically parses and analyzes data representing an image of a particular anatomical feature of a patient acquired by medical image acquisition device 40 to identify defects in the image by examining the data representing the image for predetermined patterns associated with image defects”), multiple ratings (Guehring, para 18: “An image rating processor 23 ranks individual items of the image evaluation list as mild, severe or clinically unacceptable”) related to one or more evaluation indexes of the image (Guehring, para 18: “items of the image evaluation list”; para 16: “specific classes of potential artifacts or image quality problems”);
the one or more evaluation indexes include one or more of: an anatomical structure clarity, a target portion contrast (Guehring, para 18: “contrast to noise ratio”), an image signal uniformity, an image noise level (Guehring, para 18: “SNR (signal to noise ratio)”), an artifact suppression degree (Guehring, para 18: “artifacts can be detected”), a window width (Guehring, para 15: “Field of View (FOV)”), a window level, whether the image includes a foreign object (Guehring, para 18: “automatic detection of outer body elements”), or a location accuracy of an imaged portion (Guehring, para 18: “detects anatomical landmarks for assessing correct placement of an anatomical target structure in an image”);
obtaining a quality evaluation result of the image (Guehring, see tables in Figures 5, 7A and messages/reports for the user, para 22) and displaying the quality evaluation result of the image on a terminal (Guehring, para 22: “Message and report generator 30 generates a message for presentation to a user indicating an identified defect and suggesting use of corrected image acquisition parameters for re-acquiring an image”; para 31: “A user interface generator in unit 30 automatically generates data representing a display image visually identifying at least one cause of a defect in the image and prompts a user with a resolution action or acquisition parameters to reduce the defect in the image”);
determining whether the quality evaluation result meets a preset condition (Guehring, the condition is if defects are present; para 19: “corrective action to re-acquire a defect free image”; para 21: “in response to an identified defect”);
in response to determining that the quality evaluation result does not meet the preset condition (Guehring, corrective action is suggested if there are defects, para 21), re-determining one or more current acquisition parameters (Guehring, para 21: “determine corrected image acquisition parameters for use in re-acquiring an image using image acquisition device 40 in response to an identified defect”) and displaying the one or more current acquisition parameters on the terminal (Guehring, para 31: “prompts a user with a resolution action or acquisition parameters to reduce the defect in the image”); and
adjusting at least one of a dose, an angle, an image reconstruction algorithm, or a position of the target subject based on the one or more re-determined current acquisition parameters to re-acquire an image of the target subject (Guehring, see para 21 citation above; adjusting a positioning parameter, see “solution” 519 in Figure 5: “Move table to position anatomy correctly”; para 28: “determine it is recommended to repeat an imaging scan and automatically changes appropriate imaging parameters based on determined corrective action”).
Guehring teaches one or more ratings related to one or more evaluation indexes, but fails to explicitly teach obtaining, by inputting the image of the target subject into multiple different evaluation models, multiple scores (emphasis added) related to one or more evaluation indexes of the image, wherein each of the multiple different evaluation models is configured to output a score of one of the one or more evaluation indexes of the image, the multiple different evaluation models are trained based on different training samples; and the multiple different evaluation models include a first evaluation model, a second evaluation model, a third evaluation model, a fourth evaluation model, and a fifth evaluation model, the first evaluation model is used to obtain a score of the anatomical structure clarity, the second evaluation model is used to obtain a score of the target portion contrast, the third evaluation model is used to obtain a score of the image signal uniformity, the fourth evaluation model is used to obtain a score of the image noise level, the fifth evaluation model is used to obtain a score of the artifact suppression degree. Further, Guehring teaches obtaining a quality evaluation result, as demonstrated above, but fails to explicitly teach wherein obtaining a quality evaluation result of the image is based on the multiple scores related to the one or more evaluation indexes (emphasis added).
Abdolell teaches a similar image quality system (Abdollel, abstract: “various graphical user interface tools allow a user to view various scores of certain parameters that may indicate non-conformities5 in a given image”), disclosing obtaining, by inputting the image of the target subject into multiple different evaluation models, multiple scores related to one or more evaluation indexes of the image (Abdollel, see Figure 4; para 205: “At act 402, several predictive models are selected from the plurality of predictive models and the selected predictive models use at least one image quality parameter feature as an input to determine corresponding image quality parameter scores”; see step 406 where the medical image is input, para 207: “A given image quality parameter score is determined using a predictive model that corresponds to the given image quality parameter score and uses certain image quality parameter features determined from the medical image”), wherein each of the multiple different evaluation models is configured to output a score of one of the one or more evaluation indexes of the image (Abdollel, para 207: “Each image quality parameter score may correspond to a predicted probability of the presence of the condition identified by the corresponding image quality parameter. A given image quality parameter score is determined using a predictive model that corresponds to the given image quality parameter score”), and obtaining a quality evaluation result of the image is based on the multiple scores related to the one or more evaluation indexes (Abdollel, para 209-210: “At act 408, the predicted image quality score may be determined from one or more predicted image quality parameter scores, and one or more image quality parameter features that are provided as inputs to another predictive model, which is referred to as an “overall predictive model”…At act 410, at least one of the predicted image quality score and the plurality of image quality parameter scores are output. The output can be displayed on a graphical user interface which is used to view the image”).
It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to have combined the multiple evaluation models of Abdolell with the method of Guehring in order to improve the accuracy of the different models (Abdolell, para 348: “The selection of different image/study quality parameter features for use as inputs to a predictive model to predict image/study quality parameter scores is important for improving prediction accuracy”). Guehring teaches one or more evaluation indexes above (contrast, noise, artifact, window width, foreign object, location accuracy), while Abdolell teaches using multiple different evaluation models for each individual evaluation task. Thus, the combination of Guehring in view of Abdolell teaches: the multiple different evaluation models include a first evaluation model, a second evaluation model, a third evaluation model, a fourth evaluation model, and a fifth evaluation model, the first evaluation model is used to obtain a score of the anatomical structure clarity, the second evaluation model is used to obtain a score of the target portion contrast, the third evaluation model is used to obtain a score of the image signal uniformity, the fourth evaluation model is used to obtain a score of the image noise level, the fifth evaluation model is used to obtain a score of the artifact suppression degree. Because anatomical structure clarity is not chosen as one of the one or more evaluation indexes, the first evaluation model is not required by claim 1. Additionally, FIG. 14 and para 303 of Abdolell shows at least four image quality parameters, thus teaching at least four models and scores (Abdolell, para 154: “plurality of predictive models 226 and corresponds to the particular parameter that is being assessed”).
Further, Choi teaches multiple different evaluation models that are trained based on different training samples (Choi, para 240: “The learning device can obtain a second training data set that includes a plurality of fundus images and is at least partially different from the first training data set, and train a second neural network model using the second training data set”; different models used on a target image, para 239: “training the first neural network model to classify the target fundus image into a first label or a second label”; para 241: “training a second neural network model that classifies a target fundus image into a third label or a fourth label”). It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to have trained the multiple different evaluation models on different training samples, as taught by Choi, in the method taught by Guehring in view of Abdolell in order to further improve the accuracy of the multiple evaluation models (Choi, para 397: “Multiple diagnostic auxiliary neural network models can be trained in parallel and/or independently. By training the model to predict different labels through multiple neural network models in this way, the prediction accuracy for each label can be improved and the efficiency of the prediction operation can be increased”).
Regarding claim 2 (dependent on claim 1), Guehring in view of Abdolell and Choi teaches wherein the one or more current acquisition parameters include one or more of: a portion of the target subject to be imaged, an image processing algorithm parameter, a positioning parameter, a ray source parameter, a filtering kernel parameter, an image denoising parameter, a reconstruction parameter, an artifact suppression parameter, or an injection parameter (Guehring, positioning parameter, see “solution” 519 in Figure 5: “Move table to position anatomy correctly”).
Regarding claim 3 (dependent on claim 1), Guehring in view of Abdolell and Choi teaches wherein re-determining the one or more current acquisition parameters includes: re-determining the one or more current acquisition parameters based on one or more of: an adjustment regarding an imaging condition of the target subject (Guehring, see in Figure 7A how corrective action solutions are adjustments based on imaging conditions in the leftmost column 703, “Image Artifact”), at least a part of acquisition parameters in historical imaging protocols (Guehring, para 20: “corresponding corrective image acquisition parameters”), or at least one acquisition parameter that affects at least one of the multiple scores corresponding to at least one of the one or more evaluation indexes.
Regarding claim 4 (dependent on claim 3), Guehring in view of Abdolell and Choi teaches wherein the adjustment regarding the imaging condition of the target subject includes: an adjustment regarding a positioning mode of the target subject, a removal of the foreign object, and/or an adjustment regarding the position of the target subject (Guehring, see column 709 in Figure 7A and 7B where adjustments regard positioning of the target subject, for example: “Increase Field-of-View”).
Regarding claim 6 (dependent on claim 3), Guehring in view of Abdolell and Choi teaches wherein re-determining the one or more current acquisition parameters includes re-determining the one or more current acquisition parameters based on the at least one acquisition parameter that affects the at least one of the multiple scores corresponding to the at least one of the one or more evaluation indexes (Guehring, see column 703 in Figure 7A wherein evaluation indexes are listed such as artifacts and SNR and corresponding corrective action is in the rightmost column 709; see combination with Abdolell in claim 1 regarding the multiple scores), wherein
re-determining the one or more current acquisition parameters based on the at least one acquisition parameter that affects at least one of the multiple scores corresponding to at least one of the one or more evaluation indexes includes:
determining one or more target evaluation indexes, wherein the at least one of the multiple scores corresponding to the one or more target evaluation indexes is below a preset threshold (Guehring, para 17: “list identifying items that are outside of a predetermined acceptable range (e.g. a specific type of artifact may be inevitable but becomes intrusive at a certain level determined by a threshold)”; see combination with Abdolell in claim 1 regarding the multiple scores corresponding to the evaluation indexes); and
adjusting, based on the one or more target evaluation indexes, the at least one acquisition parameter that affects the one or more target evaluation indexes (Guehring, para 28: “determine it is recommended to repeat an imaging scan and automatically changes appropriate imaging parameters based on determined corrective action”).
Regarding claim 7 (dependent on claim 1), Guehring in view of Abdolell and Choi teaches wherein the quality evaluation result of the image includes an overall quality score of the image (Abdolell, para 209: “At act 408, the predicted image quality score may be determined from one or more predicted image quality parameter scores, and one or more image quality parameter features that are provided as inputs to another predictive model, which is referred to as an “overall predictive model”), and one or more of:
a classification of the imaged portion, a disease classification of the image, or an artifact classification of the image (Guehring, para 13: “This may include (but is not limited to) number of detected defects, nature of the defect and severity. This classification can be used as parameters to query a data base to identify appropriate corrective action”; see Figures 5 and 7A where artifact classification is included, such as: “Cropped heart” in Figure 7A and “Distortion” in Figure 5; see claim 1 combination of the overall quality score of Abdolell with the quality evaluation result of Guehring).
Claims 5 and 26 are rejected under 35 U.S.C. 103 as being unpatentable over Guehring in view of Abdolell, Choi, Profio et al. (U.S. Patent No. 2018/0140270 A1), hereinafter Profio, and Xu et al. (CN Patent No. 109276269 A), hereinafter Shanghai United.
Regarding claim 5 (dependent on claim 3), Guehring in view of Abdolell and Choi teaches wherein re-determining the one or more current acquisition parameters based on at least a part of the acquisition parameters in the historical imaging protocols includes:
determining, based at least in part on posture information of the target subject (Guehring, system outputs defect information such as position of the imaged portion in the frame, see 519 in Figure 5), whether there is a matching imaging protocol in the historical imaging protocols (Guehring, para 20: “correction processor 25 uses a predetermined information map associating image defects with corresponding corrective image acquisition parameters to determine corrected image acquisition parameters for use in re-acquiring an image using image acquisition device 40 in response to an identified defect”; poor positioning is considered a defect, see para 19);
in response to determining that there is a matching imaging protocol in the historical imaging protocols, designating the at least a part of the acquisition parameters in the matching imaging protocol as the one or more current acquisition parameters (Guehring, para 28: “automatically changes appropriate imaging parameters based on determined corrective action”).
However, Guehring in view of Abdolell and Choi fails to teach 1) wherein the posture information of the target subject is based on three-dimensional contour data of the subject to be imaged, wherein the posture information of the target subject includes a relative position of the imaged portion of the subject with respect to a gantry, a thickness of the imaged portion, a width of the imaged portion, and a height of the imaged portion and 2) in response to determining that there are no matching imaging protocols in the historical imaging protocols, designating one or more default acquisition parameters as the one or more current acquisition parameters, modifying at least a portion of the one or more default acquisition parameters to obtain the one or more current acquisition parameters, or inputting one or more acquisition parameters as the one or more current acquisition parameters.
However, Profio teaches an imaging system (Profio, abstract: “Methods and systems are provided for automatically adjusting a position of a table configured to be positioned in a bore of a medical imaging device”) that utilizes posture information to determine scanning acquisition parameters (Profio, 3D patient envelope, para 42: “At 308, method 300 optionally includes identifying the patient anatomical reference and patient scan plane using patient envelope 3D imagery. That is, depth and/or visible light information acquired by the imaging sensor may be used to define the patient envelope (e.g., the surface area and volume of the patient)”), wherein the posture information of the subject includes a relative position of the imaged portion of the subject with respect to a gantry (Profio, para 42: “patient envelope (e.g., the surface area and volume of the patient), which in turn may be used to determine the position of the patient (or anatomical feature) relative to the imaging source”), a thickness of the imaged portion (Profio, para 42: “volume of the patient”; volume of the patient includes that of the imaged portion, especially wherein the entire subject is the imaged portion), a width of the imaged portion, and a height of the imaged portion (Profio, para 34: “The information received from the imaging sensor, which may include depth information and/or visible light information, may be processed to determine various subject parameters, such as subject identity, subject size (e.g., height, weight, patient envelope), and current subject position relative to the table and the imaging sensor”).
It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to have combined the posture information of Profio with the method of Guehring in view of Abdolell and Choi in order to improve the positioning accuracy of the target subject relative to the gantry based on the subject’s size (Profio, para 50: “In this way, patient positioning may be automated, decreasing setup time and the opportunity for operator error. By using the 3D patient envelope, table elevation accuracy may be improved, which may in turn improve the quality of the resulting images. Further, automating the scan setup process may allow the operator to focus more on patient care.”).
Additionally, Shanghai United teaches in response to determining that there are no matching imaging protocols in the historical imaging protocols (Shanghai United, para 41: “When the scanning protocol does not meet the requirements, the scanning protocol is optimized to obtain an updated scanning protocol”), designating one or more default acquisition parameters as the one or more current acquisition parameters, modifying at least a portion of the one or more default acquisition parameters to obtain the one or more current acquisition parameters, or inputting one or more acquisition parameters as the one or more current acquisition parameters (Shanghai United, para 42: “the scanning protocol can be optimized and the various parameter settings in the scanning protocol can be adjusted to make it meet the actual situation of the object to be scanned”). It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to have combined the response of no matching imaging protocols of Shanghai United with the method of Guehring in view of Abdolell and Choi in order to store and update previously used parameters for quick and accurate ease of use (Shanghai United, para 9: “The scanning protocol acquisition process is convenient and fast, and does not depend on the knowledge, experience and technical level of the doctor or operator”).
Regarding claim 26 (dependent on claim 5), Guehring in view of Abdolell, Choi, Profio, and Shanghai United teaches wherein determining, based at least in part on the posture information of the target subject, whether there is a matching imaging protocol in the historical imaging protocols includes: in response to that there is an imaging protocol in the historical imaging protocols corresponding to the imaged portion of the target subject and a difference in the posture information of the target subject that is within a threshold range, determining that there is a matching imaging protocol in the historical imaging protocols (Shanghai United, para 36: “quantify the difference between the information data of different scanning data groups and the information data of the object to be scanned, thereby determining the scanning data group closest to the object to be scanned… After determining the closest scan data group, the scan protocol in the scan data group can be acquired and used as the scan protocol of the object to be scanned”; information data includes posture information, body size, para 30: “the information data may include at least one of the scanned subject's age, height, weight”; see combination with Profio in claim 5);
in response to that there are no imaging protocols in the historical imaging protocols corresponding to the imaged portion of the target subject and the difference in the posture information of the target subject that is not within the threshold range, determining that there are no matching imaging protocols in the historical imaging protocols (Shanghai United, para 40: “there may be a large difference between the scanning data group in the scanning protocol database and the information data of the object to be scanned, or there may be special individual differences in the object to be scanned, resulting in the obtained scanning protocol not being a suitable choice”).
Claim 8 is rejected under 35 U.S.C. 103 as being unpatentable over Guehring in view of Abdolell, Choi, and Shanghai United.
Regarding claim 8 (dependent on claim 2), Guehring in view of Abdolell and Choi fails to teach wherein the method further comprises: if the quality evaluation result meets the preset condition, storing at least a part of the one or more current acquisition parameters as at least a part of a historical imaging protocol.
However, Shanghai United teaches if the quality evaluation result meets the preset condition, storing at least a part of the one or more current acquisition parameters as at least a part of a historical imaging protocol (Shanghai United, para 45: “after obtaining an updated scanning protocol that meets the requirements, the updated scanning protocol and the information data of the object to be scanned can be saved as a new scanning data group to the scanning protocol database”). It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to have combined the storing of acquisition parameters of Shanghai United with the method of Guehring in view of Abdolell and Choi in order to store previously used parameters for quick access of matching parameters when needed (Shanghai United, para 45: “thereby supplementing and updating the scanning protocol database to make the scanning protocol database more complete and accurate”; para 16: “The scanning protocol acquisition process is convenient and fast, and does not depend on the knowledge, experience and technical level of the doctor or operator”).
Claims 13 and 34-36 are rejected under 35 U.S.C. 103 as being unpatentable over Guehring in view of Abdolell, Choi, Profio, Shanghai United, and Fu et al. (U.S. Patent No. 2020/0160124 A1), hereinafter Fu.
Regarding claim 13 (dependent on claim 5), Guehring in view of Abdolell, Choi, Profio, and Shanghai United fails to teach wherein one of the multiple different evaluation models includes an evaluation sub-model and a feature extraction model, wherein: a feature extraction position is determined in the image of the target subject based on the feature extraction model, a local image corresponding to the image of the target subject is obtained based on the feature extraction position, and the evaluation sub-model obtains a score related to one of the one or more evaluation indexes based on one or more local features of the local image.
However, Fu teaches wherein an evaluation model includes an evaluation sub-model (Fu, sub-network 222, see FIG. 2 attached below) and a feature extraction model (Fu, sub-networks 212, 214, 219), wherein:
a feature extraction position is determined in the image of the target subject (Fu, 201 region, para 31: “sub-network 214 may determine an attention region of the image 170 based on the global feature 213 extracted by the feature extraction sub-network 212. In the example of FIG. 2, the attention region may be determined as a region 201 in the image 170”) based on the feature extraction model (Fu, feature extraction sub-networks 212 and 214, para 28: “the learning network 210 includes a sub-network 212 for extracting a feature 213 of the image 170. The sub-network 212 may also be referred to as a first sub-network or a feature extraction sub-network of the learning network 210”),
a local image corresponding to the image of the target subject is obtained based on the feature extraction position (Fu, image 202, para 32: “the region extraction section 219 may also zoom in on the attention region 201 to obtain the zoomed-in attention region (or image) 202 as the input of the learning network 220”; para 30: “sub-network 214 is used to determine a more fine-grained region in the image 170 for other sub-networks 220, such that the other sub-networks can only focus on processing in a local region at a finer scale”), and
the evaluation sub-model obtains an output related to one of the one or more evaluation indexes based on one or more local features of the local image (Fu, cropped image is input to sub-model 222, performing its goal based on the features in local image 202; para 33: “sub-network 222 performs the feature extraction at a finer scale than the full-image scale”).
PNG
media_image6.png
393
683
media_image6.png
Greyscale
It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to have combined the sub-model and feature extraction model of Fu with the method of Guehring in view of Abdolell, Choi, Profio, and Shanghai United in order to utilize finer scale features in the image to inform the evaluation score, thus improving the accuracy of the model (Fu, para 24: “In other words, accurate discriminative portion localization can promote learning fine-grained features, which in turn can further help to accurately localize the discriminative portions”; para 25: “the extraction of fine-grained features benefits from accurate localization of the discriminative portion of the object, and thus may promote the accurate recognition of the object category”). In the combination of Guehring in view of Abdolell, Choi, Profio, Shanghai United, and Fu, “the evaluation sub-model obtains a score related to one of the one or more evaluation indexes based on one or more local features of the local image” is taught. The feature extraction/sub-models architecture of the model of Fu (122 of Fu) could be used as one of the scoring models in the invention of Guehring in view of Abdolell, Choi, Profio, and Shanghai United.
Regarding claim 34 (dependent on claim 13), Guehring in view of Abdolell, Choi, Profio, Shanghai United, and Fu teaches wherein the one or more local features of the local image include color features, texture features, shape features, and local feature points (Fu, para 33: “The feature extraction sub-network 222 may extract one or more local features 223, and each local feature 223 may indicate partial feature information indicative of the image 202, such as the local color, profile, edge, line and the like of a partial object and/or partial background in the image 202”; para 1: “For example, the overall features of some birds are quite similar, and the differences only lie in the color, texture, or shape of a certain region”), and the method further includes:
determining an image quality of the local image based on the one or more local features of the local image (Fu, sub-model 222 determines an output result based on the features in the local image 202, para 30: “The region may be determined as a portion of the image that can facilitate the learning network 220 to recognize the category of the object”; image quality model output for an image taught by Guehring in claim 1, see combination with Fu in claim 13 wherein the architecture of Fu’s model may be implemented in an evaluation model); and
determining a quality of the image of the target subject based on the image quality of the local image (Fu, output of sub-model 222 determines the classification for the entire image, para 3: “determining a category of the object in the image based at least in part on the first local feature. Through this solution, it is possible to localize an image region at a finer scale accurately such that a local feature at a fine scale can be obtained for object recognition”; image quality model output for the image of the target subject taught by Guehring in claim 1, see combination with Fu in claim 13 wherein the architecture of Fu’s model may be implemented in an evaluation model).
Regarding claim 35 (dependent on claim 34), Guehring in view of Abdolell, Choi, Profio, Shanghai United, and Fu teaches wherein the feature extraction model determines the feature extraction position by generating a mask vector (Fu, region mask, para 64: “in order to ensure that the visual attention sub-network is updated or optimized in the training process, the attention region is crop on the basis that a function related with location parameter of the attention region is used as a region mask based on the visual attention sub-network”), and the mask vector is obtained by an activation function after an output of a convolutional neural network model (Fu, para 29: “a plurality of layers of a convolutional neural network (CNN) having an outstanding performance on image processing may be utilized to form the sub-network 212”; obtained by the 212 to 214 operation) and the activation function is a Rectified Linear Unit (ReLU) function (Fu, para 29: “The plurality of layers 212-1 through 212-N may further include one or more activation layers for non-linear transformation (which consists of a non-linear activation function, e.g., a ReLU function) and/or one or more pooling layers”).
Regarding claim 36 (dependent on claim 35), Guehring in view of Abdolell, Choi, Profio, Shanghai United, and Fu teaches wherein the local image is a part of the image of the target subject corresponding to the feature extraction position in the image (Fu, see 201 versus 202 in FIG. 2), and the method further includes:
obtaining the local image by multiplying corresponding elements of the mask vector based on a position correspondence relationship between the feature extraction position and the mask vector (Fu, para 66: “Based on the above Equation (3), the operation of cropping the attention region from the input image at a larger scale may be represented as element-wise multiplication between the input image and the region mask of the visual attention sub-network”).
Claims 14 and 25 are rejected under 35 U.S.C. 103 as being unpatentable over Guehring in view of Abdolell, Choi, Profio, Shanghai United, Fu, and Chen et al. (U.S. Patent No. 2020/0402229 A1), hereinafter Chen.
Regarding claim 14 (dependent on claim 13), Guehring in view of Abdollel, Choi, Profio, Shanghai United, and Fu teaches one or more scores related to the one or more evaluation indexes based on the one or more evaluation models (see claim 1 rejection), but fails to teach wherein one or more criteria for obtaining, by inputting the image of the target subject into multiple different evaluation models, the multiple scores related to the one or more evaluation indexes are different for images of the target subject acquired by scanning different scanning positions, wherein the scanning positions are body portions of the target subject corresponding to the lesion portion.
However, Chen teaches a similar method (Chen, abstract: “A quality detection unit is configured to detect an image defect in the X-ray image with regard to the type of the imaged body part and/or the type of the projection mode”) wherein one or more criteria for obtaining the one or more scores are different for images of the target subject acquired by scanning different scanning positions (Chen, para 49: “for different types of imaged body parts, the detection parameters used by the quality detection unit are different from each other; but for the same type, the detection parameters will be substantially the same”; quality detection unit influences the quality feedback output, see Figure 1 and para 31).
It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to have combined the specialized quality feedback of Chen with the method of Guehring in view of Abdollel, Choi, Profio, Shanghai United, and Fu in order to ensure that image quality is appropriately assessed based on the imaged body part, leading to more accurate feedback (Chen, para 49: “The classification or the added label should be associated with the type of imaged body part and/or the type of projection mode. Training the quality detection unit in this way makes it more sensitive to the type of the imaged body part and the type of the projection mode”).
Regarding claim 25 (dependent on claim 14), Guehring in view of Abdolell, Choi, Profio, Shanghai United, Fu, and Chen teaches wherein each of the historical imaging protocols include posture information of a historical imaged portion (Guehring, para 20: “correction processor 25 uses a predetermined information map associating image defects with corresponding corrective image acquisition parameters to determine corrected image acquisition parameters”; defect information includes position of the imaged portion in the frame, see 519 in Figure 5) and one or more types of the following information: the imaged portion (Guehring, see last citation, associated imaged portion information is saved for image defects), an image processing algorithm parameter, a positioning parameter, a radiation dose, or a projection area of a radiation beam.
Claim 21 is rejected under 35 U.S.C. 103 as being unpatentable over Shanghai United in view of Profio and Kitamura et al. (U.S. Patent No. 2014/0037194 A1), hereinafter Kitamura.
Regarding claim 21, Shanghai United teaches a method for medical image acquisition, implemented by a processing device (Shanghai United, para 58: “a medical device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the following steps when executing the program…using the scanning protocol in the scanning data group as the scanning protocol of the object to be scanned”), wherein the method includes:
obtaining posture information of a subject to be imaged, wherein the posture information of the subject includes a width of the imaged portion, and a height of the imaged portion (Shanghai United, para 32: “acquire information data of the object to be scanned”; body shape of the subject, para 34: “the information data of the object to be scanned may include basic physical characteristics of the object to be scanned, such as age, height, and weight”);
determining, based at least in part on the posture information of the subject to be imaged, whether there is a matching imaging protocol in historical imaging protocols (Shanghai United, para 36: “a query can be performed in the scanning protocol database based on the information data of the object to be scanned to query the scanning data group whose information data is closest to the object to be scanned”);
in response to determining that there is a matching imaging protocol in the historical imaging protocols, designating at least a part of acquisition parameters in the matching imaging protocol as one or more current acquisition parameters (Shanghai United, para 47: “after it is determined that the scanning protocol meets the preset requirements, the object to be scanned can be scanned and imaged according to the scanning protocol. The scanning protocol may be a scanning protocol obtained from a scanning protocol database”); and
in response to determining that there are no matching imaging protocols in the historical imaging protocols (Shanghai United, para 41: “When the scanning protocol does not meet the requirements, the scanning protocol is optimized to obtain an updated scanning protocol”), designating one or more default acquisition parameters as the one or more current acquisition parameters, modifying at least a portion of the one or more default acquisition parameters to obtain the one or more current acquisition parameters, or inputting one or more acquisition parameters as the one or more current acquisition parameters (Shanghai United, para 42: “the scanning protocol can be optimized and the various parameter settings in the scanning protocol can be adjusted to make it meet the actual situation of the object to be scanned”); and
adjusting based on the one or more re-determined current acquisition parameters to capture an image of the subject to be imaged (Shanghai United, para 30: “The scanning protocol may include various parameter settings of the medical device during the scanning imaging process, such as device model, scanning time, etc.”) based on the one or more current acquisition parameters (Shanghai United, para 47: “Specifically, after it is determined that the scanning protocol meets the preset requirements, the object to be scanned can be scanned and imaged according to the scanning protocol”).
Shanghai United fails to teach wherein the posture information of the subject is based on three-dimensional contour data of the subject to be imaged, wherein the posture information of the subject includes a relative position of an imaged portion of the subject with respect to a gantry, a thickness of the imaged portion, a width of the imaged portion, and a height of the imaged portion, wherein the obtaining the posture information of the subject to be imaged includes: collecting three-dimensional contour data of the subject to be imaged, wherein the three-dimensional contour data includes a relative distance and a relative angle between each point in the three-dimensional contour data of the subject to be imaged and an imaging device, and the relative position between the imaged portion with respect to the gantry is obtained based on the relative distance and relative angle between each point in the three-dimensional contour data of the subject to be imaged and the imaging device; and obtaining body size data of the imaged portion of the subject to be imaged based on the three-dimensional contour data of the subject to be imaged. Shanghai United further fails to explicitly teach adjusting at least one of a dose, an angle, an image reconstruction algorithm, or a position of the subject to be imaged based on the one or more re-determined current acquisition parameters to capture an image of the subject.
However, Profio teaches an imaging system (Profio, abstract: “Methods and systems are provided for automatically adjusting a position of a table configured to be positioned in a bore of a medical imaging device”) that utilizes posture information of the subject based on three-dimensional contour data of the subject to be imaged (Profio, 3D patient envelope, para 42: “At 308, method 300 optionally includes identifying the patient anatomical reference and patient scan plane using patient envelope 3D imagery. That is, depth and/or visible light information acquired by the imaging sensor may be used to define the patient envelope (e.g., the surface area and volume of the patient)”), wherein the posture information of the subject includes a relative position of an imaged portion of the subject with respect to a gantry (Profio, para 42: “patient envelope (e.g., the surface area and volume of the patient), which in turn may be used to determine the position of the patient (or anatomical feature) relative to the imaging source”), a thickness of the imaged portion (Profio, para 42: “volume of the patient”; volume of the patient includes that of the imaged portion, especially wherein the entire subject is the imaged portion), a width of the imaged portion, and a height of the imaged portion (Profio, para 34: “The information received from the imaging sensor, which may include depth information and/or visible light information, may be processed to determine various subject parameters, such as subject identity, subject size (e.g., height, weight, patient envelope), and current subject position relative to the table and the imaging sensor”), wherein the obtaining the posture information of the subject to be imaged includes: collecting three-dimensional contour data of the subject to be imaged (Profio, para 42: “At 308, method 300 optionally includes identifying the patient anatomical reference and patient scan plane using patient envelope 3D imagery. That is, depth and/or visible light information acquired by the imaging sensor”), wherein the relative position between the imaged portion with respect to the gantry is obtained based on the three-dimensional data (Profio, para 42: “patient envelope (e.g., the surface area and volume of the patient), which in turn may be used to determine the position of the patient (or anatomical feature) relative to the imaging source”); and obtaining body size data of the imaged portion of the subject to be imaged based on the three-dimensional contour data of the subject to be imaged (Profio, subject size, para 34: “The information received from the imaging sensor, which may include depth information and/or visible light information, may be processed to determine various subject parameters, such as subject identity, subject size (e.g., height, weight, patient envelope), and current subject position relative to the table and the imaging sensor”); further teaching also adjusting at least one of a dose, an angle, an image reconstruction algorithm, or a position of the subject to be imaged (Profio, adjusting the position of the target subject, para 44: “The adjustment of the table elevation may include aligning the designated patient scan plane with the scan plane or the imager and/or gantry bore, as indicated at 312”; see also para 50).
It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to have combined the three-dimensional contour body data and subsequent adjustments, taught by Profio, with the method of Shanghai United in order to improve the positioning accuracy of the target subject relative to the gantry based on the subject’s size (Profio, para 50: “In this way, patient positioning may be automated, decreasing setup time and the opportunity for operator error. By using the 3D patient envelope, table elevation accuracy may be improved, which may in turn improve the quality of the resulting images. Further, automating the scan setup process may allow the operator to focus more on patient care.”).
Additionally, Kitamura teaches a method for collecting 3D imaging data (Kitamura, abstract) wherein the three-dimensional contour data includes a relative distance and a relative angle between each point in the three-dimensional contour data of the target subject and a three-dimensional imaging device or a two-dimensional imaging device (Kitamura, para 151: “A three-dimensional laser scanner that can process three-dimensional point cloud position data will be described hereinafter. In this example, the three-dimensional laser scanner emits distance measuring light (laser light) and scans with respect to an object and measures a distance to each target point on the object therefrom based on flight time of the laser light. Then, the three-dimensional laser scanner measures the emitted direction (horizontal angle and elevation angle) of the laser light and calculates three-dimensional coordinates of the target point based on the distance and the emitted direction”). Profio specifies collecting 3D image data of the target subject, but does not specify specific methods for collecting the three-dimensional contour data. Kitamura teaches a method for utilizing a 3D scanner to collect 3D object data. Kitamura teaches a known technique of utilizing the relative distance and angle between an imaging device and the object of interest to determine the relative position of the object. A person having ordinary skill in the art, before the effective filing date of the claimed invention, could have applied the known technique, as taught by Kitamura, in the same way to the method of Shanghai United, combined with the relevant teachings of Priofio, and achieved predictable results of creating an accurate three-dimensional model to inform further image acquisition with a well-known, accessible method using a 3D camera. Doing so in the method of Shanghai United in view of Profio and Kitamura allows for obtaining the relative position between an imaged portion with respect to the gantry based on the relative distance and relative angle between each point in the three-dimensional contour data of the subject to be imaged and the three-dimensional imaging device or the two-dimensional imaging device.
Claim 24 is rejected under 35 U.S.C. 103 as being unpatentable over Guehring in view of Abdolell, Choi, Profio, Shanghai United, Fu, Chen, and Kitamura.
Regarding claim 24 (dependent on claim 14), Guehring in view of Abdolell, Choi, Profio, Shanghai United, Fu, and Chen teaches teach wherein obtaining the posture information of the target subject includes: collecting three-dimensional contour data of the subject (Profio, para 42: “At 308, method 300 optionally includes identifying the patient anatomical reference and patient scan plane using patient envelope 3D imagery. That is, depth and/or visible light information acquired by the imaging sensor”), and the relative position between the imaged portion with respect to the gantry is obtained based on the three-dimensional data (Profio, para 42: “patient envelope (e.g., the surface area and volume of the patient), which in turn may be used to determine the position of the patient (or anatomical feature) relative to the imaging source”); and obtaining body size data of the imaged portion of the target subject based on the three-dimensional contour data of the target subject (Profio, subject size, para 34: “The information received from the imaging sensor, which may include depth information and/or visible light information, may be processed to determine various subject parameters, such as subject identity, subject size (e.g., height, weight, patient envelope), and current subject position relative to the table and the imaging sensor”), but fails to explicitly teach wherein the three-dimensional contour data includes a relative distance and a relative angle between each point in the three-dimensional contour data of the target subject and a three-dimensional imaging device or a two-dimensional imaging device, and the relative position between the imaged portion with respect to the gantry is obtained based on the relative distance and relative angle between each point in the three-dimensional contour data of the target subject and the three-dimensional imaging device or the two-dimensional imaging device.
However, Kitamura teaches a method for collecting 3D imaging data (Kitamura, abstract) wherein the three-dimensional contour data includes a relative distance and a relative angle between each point in the three-dimensional contour data of the target subject and a three-dimensional imaging device or a two-dimensional imaging device (Kitamura, para 151: “A three-dimensional laser scanner that can process three-dimensional point cloud position data will be described hereinafter. In this example, the three-dimensional laser scanner emits distance measuring light (laser light) and scans with respect to an object and measures a distance to each target point on the object therefrom based on flight time of the laser light. Then, the three-dimensional laser scanner measures the emitted direction (horizontal angle and elevation angle) of the laser light and calculates three-dimensional coordinates of the target point based on the distance and the emitted direction”). Profio specifies collecting 3D image data of the target subject, but does not specify specific methods for collecting the three-dimensional contour data. Kitamura teaches a method for utilizing a 3D scanner to collect 3D object data. Kitamura teaches a known technique of utilizing the relative distance and angle between an imaging device and the object of interest to determine the relative position of the object. A person having ordinary skill in the art, before the effective filing date of the claimed invention, could have applied the known technique, as taught by Kitamura, in the same way to the method of Guehring, combined with the relevant teachings of Priofio in the combination of Guehring in view of Abdolell, Choi, Profio, Shanghai United, Fu, and Chen, and achieved predictable results of creating an accurate three-dimensional model to inform further image acquisition with a well-known, accessible method using a 3D camera. Doing so allows for obtaining the relative position between an imaged portion with respect to the gantry based on the relative distance and relative angle between each point in the three-dimensional contour data of the subject to be imaged and the three-dimensional imaging device or the two-dimensional imaging device.
Claim 28 is rejected under 35 U.S.C. 103 as being unpatentable over Guehring in view of Abdollel, Choi, and Nye et al. (U.S. Patent No. 2019/0150857 A1), hereinafter Nye.
Regarding claim 28 (dependent on claim 1), Guehring in view of Abdollel and Choi teaches wherein the one or more evaluation indexes include the target portion contrast, the window width and/or the window level of the image (see claim 1 rejection), but fails to teach wherein the method further comprises: when the target portion contrast, the window width and/or the window level of the image meets a preset threshold condition; when there is no foreign object in the image; and when the imaged portion is located in an imaging area, determining that the quality evaluation result meets the preset condition.
However, Nye teaches when the target portion contrast, the window width and/or the window level of the image meets a preset threshold condition (Nye, determines if there is sufficient contrast, para 118: “Other quality control checks can include an evaluation of sufficient contrast, an analysis of a level of noise or artifact in the image, an examination of appropriate/sufficient dosage for image clarity, etc”);
when there is no foreign object in the image (Nye, para 120: “If analysis is to proceed (e.g., because the image passes quality check(s) and/or an instruction indicates to proceed despite image quality concerns, etc.), then, at 1214, the image data is evaluated with respect to a clinical check… identify a severe pneumothorax and/or other condition (e.g., tube within the right mainstem, free air in the bowel, fracture, tumor, lesion, other foreign object, etc.) in the image data. If no finding is determined, then the process 1200 ends”); and
when the imaged portion is located in an imaging area (Nye, para 118: “the image data is analyzed to determine whether the associated image has good patient positioning (e.g., the patient is positioned such that an anatomy or region of interested is centered in the image, etc.)”), determining that the quality evaluation result meets the preset condition (Nye, see Figure 12 wherein there are a series of quality control checks to determine if image satisfies various requirements).
It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to have combined the quality requirements of Nye with the method of Guehring in view of Abdolell and Choi in order to improve medical imaging by requiring that foreign objects are removed and positioning is correct (Nye, para 138: “From the foregoing, it will be appreciated that the above disclosed methods, apparatus, and articles of manufacture have been disclosed to monitor, process, and improve operation of imaging and/or other healthcare systems using a plurality of deep learning and/or other machine learning techniques”).
Claim 38 is rejected under 35 U.S.C. 103 as being unpatentable over Guehring in view of Abdolell, Choi, and Hao et al. (U.S. Patent No. 10,467,507 B1), hereinafter Hao.
Regarding claim 38 (dependent on claim 1), Guehring in view of Abdolell and Choi teaches wherein the obtaining the quality evaluation result of the image based on the multiple scores related to the one or more evaluation indexes and displaying the quality evaluation result of the image on a terminal includes: obtaining an overall quality score of the target image by summing up and averaging the multiple scores of the one or more evaluation indexes (Abdolell, para 218: “a consensus label is determined from the plurality of image quality parameter inputs. This consensus may be mathematically determined, for example by using the mean”), but fails to teach determining weights to the one or more evaluation indexes, wherein the weights are given by a trained weight model, an input of the trained weight model includes the multiple scores related to the one or more evaluation indexes and the target image, and an output of the trained weight model includes the weights to the one or more evaluation indexes; and averaging after multiplying the weights by the multiple scores.
However, Hao teaches determining weights to the one or more evaluation indexes (Hao, col 9, ln 50-51: “image score weight values for each composition rule”), wherein the weights are given by a trained weight model (Hao, col 9, ln 50: “composition rule extraction module 212”), an input of the trained weight model includes the multiple scores related to the one or more evaluation indexes and the target image (Hao, col 8, ln 9-14: “The image scoring module 120 may include the composition rule extraction module 212 for extracting image classifying features from images and training one or more machine learning models for determining a score or classification of the image (e.g., good or bad quality image, etc.)”), and an output of the trained weight model includes the weights to the one or more evaluation indexes (Hao, col 9, ln 50-56: “For example, a first composition rule corresponding to busy background may have stronger influence than a second composition rule of centering the object in the image, thus the first weight value for the first composition rule may be higher than the second weight value for the second composition rule”); and obtaining an overall quality score of the target image after multiplying the weights by the multiple scores of the one or more evaluation indexes (Hao, col 9, ln 56-60: “Accordingly, an image score may be based on a composition score that is based at least in part on the first composition rule adjusted by the first weight value and the second composition rule adjusted by the second weight value.”).
It would have been obvious to a person having ordinary skill in the art, before the effective filing date of the claimed invention, to have combined the trained weight model of Hao with the method of Guehring in view of Abdolell and Choi in order to appropriately adjust the influence of certain types of scores based on their relevance to the subject of the image (Hao, col 9, ln 50-56 – cited above). In the combination of Guehring in view of Abdolell, Choi, and Hao, weight values are determined and applied to the mean operation for determining an overall score value.
Allowable Subject Matter
Claim 37 would be allowable if 1) rewritten to overcome the rejection(s) under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), 2nd paragraph, set forth in this Office action and 2) rewritten in independent form including all of the limitations of the base claim and any intervening claims. The following is a statement of reasons for the indication of allowable subject matter: regarding claim 37, Shanghai United teaches querying a scanning protocol database for the closest information data, but fails to teach wherein determining whether there is a matching imaging protocol includes both 1) the specific search keywords recited in claim 37 and 2) wherein the information data/posture information (introduced in independent claim 21) is compared to a difference threshold range. Profio, Kitamura, and similar prior art fail to remedy these deficiencies.
Thus, the prior art fails to teach as a whole wherein the determining, based at least in part on the posture information of the subject to be imaged, whether there is the matching imaging protocol in the historical imaging protocols includes:
searching the historical imaging protocols based on search keywords, and selecting several historical imaging protocols as candidate imaging protocols, wherein the search keywords include positioning information of the subject to be imaged in the imaging protocol, the positioning information includes standing imaging, side lying imaging, and supine imaging, and the candidate imaging protocols match the search keywords;
in response to there is an imaging protocol in the candidate imaging protocols that is the same as the imaged portion and whose posture information has a difference with the posture information of the subject to be imaged that is within a threshold range, determining that there is a matching imaging protocol in the historical imaging protocols; or in response to there is no imaging protocol in the candidate imaging protocols that is the same as the imaged portion and whose posture information has a difference with the posture information of the subject to be imaged that exceeds the threshold range, determining that there is no matching imaging protocol in the historical imaging protocols.
In view of the foregoing, the prior art references alone or in reasonable combination are insufficient to teach the invention as a whole, as claimed in claim 37.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Booij et al. (Booij, R., van Straten, M., Wimmer, A., & Budde, R. P. (2018). Automated patient positioning in CT using a 3D camera for body contour detection: accuracy in pediatric patients. European radiology, 31(1), 131-138.) teaches utilizing body contour data to position a target subject for CT imaging.
Liu et al. (Liu, T. J., Lin, W., & Kuo, C. C. J. (2012). Image quality assessment using multi-method fusion. IEEE Transactions on image processing, 22(5), 1793-1807.) teaches an image quality assessment method combining multiple different scores (abstract: “The research is motivated by the observation that there is no single method that can give the best performance in all situations. To achieve MMF, we adopt a regression approach. The new MMF score is set to be the nonlinear combination of scores from multiple methods with suitable weights obtained by a training process”).
Any inquiry concerning this communication or earlier communications from the examiner should be directed to EMMA E DRYDEN whose telephone number is (571)272-1179. The examiner can normally be reached M-F 9-5 EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, ANDREW BEE can be reached at (571) 270-5183. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/EMMA E DRYDEN/Examiner, Art Unit 2677
/ANDREW W BEE/Supervisory Patent Examiner, Art Unit 2677