DETAILED ACTION
Claim Status
This action is in response to the applicant’s arguments filed on February 17, 2026. Claims 1-20 are pending for examination in this application.
Response to Arguments
Applicant's arguments filed February 17, 2026, have been fully considered but they are not persuasive.
Argument: On page 7, the applicant alleges, “The combination of Eusemann, Mailhe, and Vitsnudel fails to teach or suggest the claimed scoring. Because Vitsnudel does not teach determining a score based on the network's performance accuracy against a gold standard for a clinical task, the reference fails to teach the "scoring" limitation as it would be understood in the context of the claimed invention.”
Response: The examiner respectfully disagrees. The applicant’s Specification states scoring an output in varying ways, in Paragraph [0008], “Scoring comprises comparing the output to expert annotated data,” in Paragraph [0044], “The outputs of the networks 230 are then judged against an expertly annotated output (that, for example, may be agreed to be the optimal or correct output),” and Paragraph [0051], “The score may represent the difference between the output and gold standard annotations / calculations.” The scoring limitation given the broadest reasonable interpretation in light of the Specification is interpreted as comparing the output of the two identical machine trained networks and comparing them to a reference. Vitsnudel teaches, scoring an output of each of the identical machine trained networks for each set of the input plurality of sets of imaging data, see Vitsnudel, Fig. 3, Neural Network Instance 320(1) and Neural Network Instance 320(2) and Col 7, Lines, 65-67 – Col 8, Lines 1-18, “Neural network instance 320(1) determines, from feature set 322(1), what output value 324(1) to output according to its network, which is defined by parameters in neural network data store 104 … Neural network instance 320(2) determines, from feature set 322(2), what output value 324(2) to output according to its network, which is also defined by parameters in neural network data store 104. In this manner, a “siamese neural network” is implemented,” Siamese neural networks are designed to compare two inputs through identical subnetworks and compare the outputs, and Vitsnudel Col 8, Lines 17-28, “A comparator 326 compares output value 324(1) and output value 324(2) to determine which is greater (or which is lesser) and outputs a relative metric 328 … an input image is compared to a plurality of representative images and a relative metric determined for each comparison. Since the images in RDS 110 have corresponding known metrics 308, those can be stored with the comparisons in match results data store 112,” comparing the output values and comparing to known metrics is considered to be scoring an output of each of the identical machine trained networks for each set of the input plurality of sets of imaging data.
Argument: On page 8, the applicant alleges, “The combination of Eusemann, Mailhe, and Vitsnudel fails to teach or suggest the “separate identical networks” as claimed. The claimed method uses identical networks as a consistent benchmark to evaluate multiple different reconstructions that are all derived from the same single set of raw data in a single, unified analysis.”
Response: The examiner respectfully disagrees. Mailhe teaches, inputting each set of the plurality of sets of imaging data into separate identical machine trained networks for the given clinical task in Paragraph [0091], “The same deep machine-learned network may be used for different patients. The same or different copies of the same machine-learned network are applied for different patients, resulting in reconstruction of patient-specific representations or reconstructions using the same values or weights of the learned parameters of the network. Different patients and/or the same patient at a different time may be scanned while the same or fixed trained network is used in reconstruction the image,” the same patient scanned at a different time is considered to be plurality of sets of imaging data, using different copies of the same machine-learned network is considered to be inputting each set of the plurality of sets of imaging data into separate identical machine trained networks for the given clinical task that are all derived from the same single set of raw data in a single unified analysis.
Argument: On page 8, the applicant alleges, “The combination of Eusemann, Mailhe, and Vitsnudel fails to teach or suggest the identifying optimal reconstruction parameters based on a comparison of scores. The Applicants respectfully disagree with this interpretation as the "parameters" being optimized in Mailhe are fundamentally different from the "parameters" being identified in the Applicant's specification and claims”
Response: The examiner respectfully disagrees. Mailhe was relied on to teach, and identifying optimal in Paragraph [0042], “Machine learning is an offline training phase where the goal is to identify an optimal set of values of learnable parameters of the model that can be applied to many different inputs (i.e., image domain data after gradient calculation in the optimization or minimization of the reconstruction),” identify an optimal set of values of learnable parameters of a model that can be applied to many different inputs i.e., image domain data after gradient calculation in the optimization of the reconstruction is considered to be determining which input image created by certain reconstruction parameters are best; Eusemann teaches the reconstruction parameters. Vitsnudel was relied on to teach by comparing the scores for each set of the input plurality of sets of imaging data in Col 8, Lines 17-28, “A comparator 326 compares output value 324(1) and output value 324(2) to determine which is greater (or which is lesser) and outputs a relative metric 328 … an input image is compared to a plurality of representative images and a relative metric determined for each comparison. Since the images in RDS 110 have corresponding known metrics 308, those can be stored with the comparisons in match results data store 112,” comparing the output values and comparing to known metrics is considered to be scoring an output of each of the identical machine trained networks for each set of the input plurality of sets of imaging data.
Argument: On pages 10-11, the applicant alleges, “The motivation to combine Eusemann and Mailhe is conclusory.”
Response: The examiner respectfully disagrees. On pages 6-7 of the Non-Final Office Action mailed on December 02, 2025. Maile was relied on just teaching, inputting each set of the plurality of sets of imaging data into separate identical machine trained networks for the given clinical task in Paragraph [0091], “The same deep machine-learned network may be used for different patients. The same or different copies of the same machine-learned network are applied for different patients, resulting in reconstruction of patient-specific representations or reconstructions using the same values or weights of the learned parameters of the network. Different patients and/or the same patient at a different time may be scanned while the same or fixed trained network is used in reconstruction the image,” the same patient scanned at a different time is considered to be plurality of sets of imaging data, using different copies of the same machine-learned network is considered to be inputting each set of the plurality of sets of imaging data into separate identical machine trained networks for the given clinical task that are all derived from the same single set of raw data in a single unified analysis; and identifying optimal in Mailhe, Paragraph [0042], “Machine learning is an offline training phase where the goal is to identify an optimal set of values of learnable parameters of the model that can be applied to many different inputs (i.e., image domain data after gradient calculation in the optimization or minimization of the reconstruction)”.
Thus it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to implement the teachings of Mailhe into Eusemann because by utilizing means of Mailhe to use data of the same patient at different times as input into copies of the same machine-trained network in the method of Eusemann utilizing three dimensional images generated based on different reconstruction techniques because the concept of identifying an optimal set of values of parameters that can be applied to many different inputs (i.e., image domain data after gradient calculation in the optimization or minimization of the reconstruction) would be obvious to one of ordinary skill in the art to use copies of the same machine-learned network to process the three dimensional images generated based on different reconstruction techniques of Eusemann. Thus it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to implement the teachings of Mailhe into Eusemann to improve the speed and reconstructive quality. Therefore, the combination of the references teaches the limitation.
Argument: On page 11, the applicant alleges, “The motivation to combine Vitsnudel with Eusemann and Mailhe is flawed and based on an inapplicable benefit. The rejection's purported benefit for incorporating Vitsnudel's method is the efficiency gain from pre-processing a "representative image set. "However, the claimed invention does not use, require, or even mention a "representative image set." The claim requires comparing the scores generated from the "plurality of sets of reconstructed imaging data." This is a self-contained comparison where a limited number of reconstructions (e.g., 10 or 100) are evaluated against each other or a gold standard to find the best among them.”
Response: The examiner respectfully disagrees. The scoring limitation given the broadest reasonable interpretation in light of the Specification is interpreted as comparing the output of the two identical machine trained networks and comparing them to a reference. Vitsnudel teaches, by comparing the scores for each set of the input plurality of sets of imaging data, see Vitsnudel Col 8, Lines 17-28, “A comparator 326 compares output value 324(1) and output value 324(2) to determine which is greater (or which is lesser) and outputs a relative metric 328 … an input image is compared to a plurality of representative images and a relative metric determined for each comparison. Since the images in RDS 110 have corresponding known metrics 308, those can be stored with the comparisons in match results data store 112,” comparing the output values and comparing to known metrics is considered to be scoring an output of each of the identical machine trained networks for each set of the input plurality of sets of imaging data.
Argument: On page 12, the applicant alleges, “The combination of Eusemann and Cheng fails to teach training a plurality of networks as claimed. In contrast, the claim requires training different, independent networks on different types of reconstruction data (e.g., Network A trained on "sharp kernel" images from Eusemann, Network B on "soft kernel" images). The goal is to then compare these networks in a parallel process to see which one performs a clinical task better. Cheng teaches a serial, cascaded training process for acceleration. The claim requires a parallel, comparative training process for optimization.”
Response: The examiner respectfully disagrees. The claim limitations as written do not explicitly state, “a parallel, comparative training process for optimization.” Therefore, Cheng, teaches, generating a plurality of different trained networks by machine training an original network, wherein each different trained network of the plurality of different trained networks uses a different set of the plurality of sets of imaging data; in Paragraph [0056], “it is possible to train a series of neural networks in this fashion, such as different models or networks for different stages of an iterative reconstruction process. For example, a first neural network may be trained by using early estimates (I.sub.m1, I.sub.m2, and I.sub.m3) as input and an estimate I.sub.N1 at iteration number N1 as the first target image where N1>m3. A second neural network may then be trained using I.sub.N1 as an input and an estimate I.sub.N2 at iteration N2 as the target image, where N2>N1. These sub-networks can be cascaded to form one deeper neural network and serve to pre-train and initialize a final, deeper network, so that the final network can achieve faster convergence and better performance,” different stages of an iterative reconstruction process are considered to be a plurality of sets of imaging data and a first neural network trained with early estimates and a second neural network trained with another set of estimates, where the sub-networks can be cascaded is considered to be generating a plurality of different trained networks by machine training an original network.
Argument: On page 13, the applicant alleges, “The combination of Eusemann and Cheng fails to teach comparing the performance of different networks. Claim 10, in contrast, requires comparing the performance of multiple, different trained networks against each other to determine which network is superior for the clinical task.”
Response: The examiner respectfully disagrees. Cheng teaches, comparing a performance of the plurality of different trained networks for the given clinical task in Paragraph [0028], “the deep learning algorithms may process (either in a supervised or guided manner or in an unsupervised or unguided manner) the known or training data sets until the mathematical relationships between the initial data and desired output(s) are discerned and/or the mathematical relationships between the inputs and outputs of each layer are discerned and characterized. Similarly, separate validation data sets may be employed in which both input and desired target values are known, but only the initial values are supplied to the trained deep learning algorithms, with the outputs then being compared to the outputs of the deep learning algorithm to validate the prior training and/or to prevent over-training”, by comparing the mathematical relationships between the inputs and outputs with input and desired target value that is considered to be the comparing a performance; this process may be done on multiple neural networks.
Argument: On page 13, the applicant alleges, “The combination of Eusemann and Cheng fails to teach selecting the optimized network. The claim requires selecting the optimized machine trained network from a plurality of different networks.”
Response: The examiner respectfully disagrees. In the applicant’s Specification, Paragraph [0058], “At act 840, the system 200 selects the optimized machine trained network 230 based the comparison. The ground truth / annotations are compared to the output to determine the performance of the machine trained networks 230. The best performing, e.g., the network 230 that generates an output that is most similar to the annotations may be selected as the optimal machine trained network 230 for the particular task.” Cheng teaches and selecting the optimized machine trained network based the comparison in Paragraph [0047], where I.sub.max corresponds to the optimal result where the cost function defined by the depicted curve is satisfied (here, maximized).”, I.sub.max is considered to be optimal reconstruction parameters in which would be compared to annotated data; Comparing the outputs was stated in argument above, the optimized machine trained network based on the comparison may be selected.
Argument: On page 14, the applicant alleges, “The stated motivation to combine Eusemann and Cheng is insufficient.”
Response: The examiner respectfully disagrees. Cheng was relied on just teaching “generating a plurality of different trained networks by machine training an original network, wherein each different trained network of the plurality of different trained networks uses a different set of the plurality of sets of imaging data; comparing a performance of the plurality of different trained networks for the given clinical task; and selecting the optimized machine trained network based the comparison” Cheng Paragraphs [0056], [0028], and Paragraph [0047]. Thus it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to implement the teachings of Cheng into Eusemann because by utilizing means of Cheng to train a series of neural networks such as different models or networks for different stages of an iterative process and validate the performance in to the method of Eusemann utilizing three dimensional images generated based on different reconstruction techniques because the concept of training a series of neural networks for different iterative reconstruction stages which is considered to be different reconstruction parameters would be obvious to one of ordinary skill in the art to train a series of neural networks to process the three dimensional images generated based on different reconstruction techniques of Eusemann. Thus it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to implement the teachings of Cheng into Eusemann to provide improved benefits, such as increased reconstruction efficiency or speed, while still achieving good image quality or allowing a low patient dose, achieving good image quality is important in Eusemann as generating multiple reconstructions and determining the optimal parameters is based on comparison to a reference image.
Argument: On page 15, the applicant alleges, “Eusemann fails to teach selecting parameters as claimed. The specification makes it clear that the "optimal" parameters are identified through an evaluative process of generating multiple reconstructions, scoring them against a ground truth, and comparing the scores to find the best performer (as recited in the method of claim 1).”
Response: The examiner respectfully disagrees. The limitations of claim 17 does not explicitly state “scoring an output of each of the identical machine trained networks for each set of the input plurality of sets of imaging data; and identifying optimal reconstruction parameters by comparing the scores for each set of the input plurality of sets of imaging data,” as recited in the method of Claim 1. The scoring and comparison recited in claim 1 is not required in the context of Claim 17 as it does not invoke means" (or "step") with functional language.
Argument: On page 15, the applicant alleges, “The rejection improperly relies on an incompatible embodiment. The claimed concept involves taking a single set of raw data and creating a plurality of different reconstructions from it to find the optimal set of parameters (Specification, Figure 8, Act 810). The image processor of claim 17 is configured to perform this bifurcated reconstruction from a single source.”
Response: The examiner respectfully disagrees. Eusemann teaches the limitations of claim 17 as written and an image processor configured to select first reconstruction parameters for the machine trained network for machine learning detection (see Eusemann Fig. 6, S620 Acquire next set of image data using other imaging parameters and Paragraph [0053], “At S620, it may be determined to acquire a second set of image data using a second set of imaging parameters which are suited for a particular reconstruction technique in order to generate a three-dimensional image which is optimized for a first automated feature extraction system,” a second set of imaging parameter is considered to be select first reconstruction parameters for the machine trained network for machine learning detection and Paragraph [0054], “Accordingly, each organ/disease pair may be associated with multiple sets of imaging parameters (i.e., with each set of imaging parameters corresponding to a respective row of structure 700), and each set of imaging parameters may in turn be associated with one or more reconstruction techniques. Two or more different sets of imaging parameters may be associated with a same reconstruction technique, and two or more different reconstruction techniques may be associated with a same set of imaging parameters,” each set of imaging parameters may be associated with one or more reconstruction techniques therefore, a single set of imaging parameters may have two reconstruction techniques that would be considered to be a first reconstruction parameter and a second reconstruction parameter)
and second reconstruction parameters for human consumption (see Eusemann, Fig. 6, S610 Acquire a set of image data of a patient volume using first imaging parameters, Paragraph [0053], “the first set of imaging parameters used at S610 may be suited for a particular reconstruction technique in order to generate a three-dimensional image which is optimized for human identification of features,” the first set of imaging parameters is considered to be second reconstruction parameters for human consumption and Paragraph [0054], “Accordingly, each organ/disease pair may be associated with multiple sets of imaging parameters (i.e., with each set of imaging parameters corresponding to a respective row of structure 700), and each set of imaging parameters may in turn be associated with one or more reconstruction techniques. Two or more different sets of imaging parameters may be associated with a same reconstruction technique, and two or more different reconstruction techniques may be associated with a same set of imaging parameters,” each set of imaging parameters may be associated with one or more reconstruction techniques therefore, a single set of imaging parameters may have two reconstruction techniques that would be considered to be a first reconstruction parameter and a second reconstruction parameter).
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 17-18, and 20 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Eusemann et al, US 20190046129.
Regarding claim 17, Eusemann teaches a system
for clinical aided diagnostics of a given clinical task, the system comprising (see Eusemann, Paragraph [0039], “The one or more images may be generated using image processing techniques which are intended to generate images suitable for optimized and automated feature extraction using computer-aided diagnosis systems. For example, the one or more images may comprise three-dimensional images which are generated using image reconstruction techniques intended to generate three-dimensional images suitable for optimized and automated feature extraction using computer-aided diagnosis systems,” computer-aided diagnosis systems is considered to be clinical aided diagnostics given a clinical task):
a medical imaging device configured to acquire raw data (see Eusemann, Paragraph [0015]-[0016], “Imaging system 110 comprises a CT scanner including X-ray source 111 for emitting X-ray beam 112 toward opposing radiation detector 113. … Detector 113 receives the radiation and produces a set of data (i.e., a raw image) for each projection angle.”, imaging system 110 is considered to be a medical imaging device);
a machine trained network configured for the given clinical task (see Eusemann, Paragraph [0049], “Similarly, the one or more images generated at S230 may be generated by inputting the one or more images to a trained network in order generate one or more images which are better-suited to automated feature extraction.”, the one or more images are acquired from the given clinical task and input into a trained network);
and an image processor configured to select first reconstruction parameters for the machine trained network for machine learning detection (see Eusemann Fig. 6, S620 Acquire next set of image data using other imaging parameters and Paragraph [0053], “At S620, it may be determined to acquire a second set of image data using a second set of imaging parameters which are suited for a particular reconstruction technique in order to generate a three-dimensional image which is optimized for a first automated feature extraction system,” a second set of imaging parameter is considered to be select first reconstruction parameters for the machine trained network for machine learning detection and Paragraph [0054], “Accordingly, each organ/disease pair may be associated with multiple sets of imaging parameters (i.e., with each set of imaging parameters corresponding to a respective row of structure 700), and each set of imaging parameters may in turn be associated with one or more reconstruction techniques. Two or more different sets of imaging parameters may be associated with a same reconstruction technique, and two or more different reconstruction techniques may be associated with a same set of imaging parameters,” each set of imaging parameters may be associated with one or more reconstruction techniques therefore, a single set of imaging parameters may have two reconstruction techniques that would be considered to be a first reconstruction parameter and a second reconstruction parameter)
and second reconstruction parameters for human consumption (see Eusemann, Fig. 6, S610 Acquire a set of image data of a patient volume using first imaging parameters, Paragraph [0053], “the first set of imaging parameters used at S610 may be suited for a particular reconstruction technique in order to generate a three-dimensional image which is optimized for human identification of features,” the first set of imaging parameters is considered to be second reconstruction parameters for human consumption and Paragraph [0054], “Accordingly, each organ/disease pair may be associated with multiple sets of imaging parameters (i.e., with each set of imaging parameters corresponding to a respective row of structure 700), and each set of imaging parameters may in turn be associated with one or more reconstruction techniques. Two or more different sets of imaging parameters may be associated with a same reconstruction technique, and two or more different reconstruction techniques may be associated with a same set of imaging parameters,” each set of imaging parameters may be associated with one or more reconstruction techniques therefore, a single set of imaging parameters may have two reconstruction techniques that would be considered to be a first reconstruction parameter and a second reconstruction parameter),
reconstruct a first image using the first reconstruction parameters (Fig. 6 and Paragraph [0056], “The one or more three-dimensional images are generated at S650 using image reconstruction techniques which are intended to generate three-dimensional images suitable for automated feature extraction using computer-aided diagnosis systems”),
and input the first image into the machine trained network for the given clinical task (see Eusemann, Fig. 6 and Paragraph [0056], “Automated feature extraction is performed at S670 based on the one or more three-dimensional images generated at S650,” automated feature extraction is performed is considered to be input the first image into the machine trained network),
the image processor further configured to reconstruct a second image using the second reconstruction parameters (see Eusemann, Fig. 6, and Paragraph [0055], “The three-dimensional image may be generated using any three-dimensional reconstruction technique which generates a three-dimensional image suitable for human viewing,”);
wherein the second image and the output of the machine trained network are provided to an operator (see Eusemann Fig. 6 and Paragraph [0055], “A two-dimensional slice image of the image generated at S640 may then be displayed to an operator on terminal 130 and/or on another display of a separate computing system.).
Regarding claim 18, Eusemann teaches the system of claim 17, further comprising:
a display configured to display the second image the output of the machine trained network (see Eusemann Fig. 6 and Paragraph [0055], “A two-dimensional slice image of the image generated at S640 may then be displayed to an operator on terminal 130 and/or on another display of a separate computing system.”).
Regarding claim 20, Eusemann further teaches the system of claim 17,
wherein the medical imaging device comprises one of a CT device, MRI device, X-ray device, or ultrasound device (see Eusemann, Paragraph [0015], “Imaging system 110 comprises a CT scanner including X-ray source 111 for emitting X-ray beam 112 toward opposing radiation detector 113”).
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.
Claim(s) 1, 6, and 8-9 is/are rejected under 35 U.S.C. 103 as being unpatentable over Eusemann et al, US 20190046129 in view of Mailhe et al, US 20200408864 in view of Vitsnudel et al, US 10210627.
Regarding claim 1, Eusemann teaches a method for
determining optimal reconstruction parameters for clinical aided diagnostics of a given clinical task, the method comprising (see Eusemann, Paragraph [0039], “The one or more images may be generated using image processing techniques which are intended to generate images suitable for optimized and automated feature extraction using computer-aided diagnosis systems. For example, the one or more images may comprise three-dimensional images which are generated using image reconstruction techniques intended to generate three-dimensional images suitable for optimized and automated feature extraction using computer-aided diagnosis systems,” computer-aided diagnosis systems is considered to be clinical aided diagnostics given a clinical task):
generating a plurality of sets of reconstructed imaging data from a set of raw data, each set of the plurality of sets of imaging data using unique combinations of reconstruction parameters (see Eusemann, Fig 6, S650, For each set of image data, generate one or more three-dimensional images for automated feature extraction, and Paragraph [0050], “process 600 utilizes multiple sets of image data, with each set of image data being acquired using different imaging parameters” and Paragraph [0054], “Two or more different sets of imaging parameters may be associated with a same reconstruction technique, and two or more different reconstruction techniques may be associated with a same set of imaging parameters”);
inputting each set of the plurality of sets of imaging data into (see Eusemann, Paragraph [0049], “Similarly, the one or more images generated at S230 may be generated by inputting the one or more images to a trained network in order generate one or more images which are better-suited to automated feature extraction.”);
Eusemann does not expressively teach
inputting each set of the plurality of sets of imaging data into separate identical machine trained networks for the given clinical task;
and identifying optimal reconstruction parameters by comparing the scores for each set of the input plurality of sets of imaging data.
However, Mailhe in a similar invention in the same field of endeavor teaches
inputting each set of the plurality of sets of imaging data into separate identical machine trained networks for the given clinical task (see Mailhe, Paragraph [0091], “The same or different copies of the same machine-learned network are applied for different patients, resulting in reconstruction of patient-specific representations or reconstructions using the same values or weights of the learned parameters of the network,” different copies of the same machine-learned network are applied for different patients is considered to be inputting each set of the plurality of sets of imaging data into separate identical machine trained networks);
and identifying optimal (see Mailhe, Paragraph [0042], “Machine learning is an offline training phase where the goal is to identify an optimal set of values of learnable parameters of the model that can be applied to many different inputs (i.e., image domain data after gradient calculation in the optimization or minimization of the reconstruction),” identify an optimal set of values of learnable parameters of a model that can be applied to many different inputs i.e., image domain data after gradient calculation in the optimization of the reconstruction is considered to be determining which input image created by certain reconstruction parameters are best; Eusemann teaches the reconstruction parameters.)
The combination of Eusemann and Mailhe are analogous art as they are both in the same field of endeavor of reconstruction for optimizing medical images. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, for different copies of the same machine-learned network applied to be applied for different patients as taught in the method of Mailhe in the method of Eusemann for patient-specific representations or reconstructions using the same values or weights of the learned parameters of the network (see Mailhe, Paragraph [0091]).
Eusemann in view of Mailhe does not expressively teach
scoring an output of each of the identical machine trained networks for each set of the input plurality of sets of imaging data
by comparing the scores for each set of the input plurality of sets of imaging data
However, Vitsnudel in a similar invention in the same field of endeavor teaches
scoring an output of each of the identical machine trained networks for each set of the input plurality of sets of imaging data (see Vitsnudel, Fig. 3, Neural Network Instance 320(1) and Neural Network Instance 320(2) and Col 7, Lines, 65-67 – Col 8, Lines 1-18, “Neural network instance 320(1) determines, from feature set 322(1), what output value 324(1) to output according to its network, which is defined by parameters in neural network data store 104 … Neural network instance 320(2) determines, from feature set 322(2), what output value 324(2) to output according to its network, which is also defined by parameters in neural network data store 104. In this manner, a “siamese neural network” is implemented,” Siamese neural networks are designed to compare two inputs through identical subnetworks and compare the outputs, and Vitsnudel Col 8, Lines 17-28, “A comparator 326 compares output value 324(1) and output value 324(2) to determine which is greater (or which is lesser) and outputs a relative metric 328 … an input image is compared to a plurality of representative images and a relative metric determined for each comparison. Since the images in RDS 110 have corresponding known metrics 308, those can be stored with the comparisons in match results data store 112,” comparing the output values and comparing to known metrics is considered to be scoring an output of each of the identical machine trained networks for each set of the input plurality of sets of imaging data);
by comparing the scores for each set of the input plurality of sets of imaging data (see Vitsnudel, Col 5, Lines 28-32, “when an input image is processed, an output score is determined and then compared over the scores of the images of the representative image set,” and see Vitsnudel Col 8, Lines 17-28, “A comparator 326 compares output value 324(1) and output value 324(2) to determine which is greater (or which is lesser) and outputs a relative metric 328 … an input image is compared to a plurality of representative images and a relative metric determined for each comparison. Since the images in RDS 110 have corresponding known metrics 308, those can be stored with the comparisons in match results data store 112,” comparing the output values and comparing to known metrics is considered to be scoring an output of each of the identical machine trained networks for each set of the input plurality of sets of imaging data).
The combination of Eusemann Mailhe, and Vitsnudel are analogous art as they are all in the same field of endeavor of training a neural network and determining metrics of the input image. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, for respective neural network instances 320(1) and 320(2) are implemented as a siamese neural network; output a score when an input image is processed and compare the scores of the images as taught in the computer system of Vitsnudel in the method of Eusemann in view of Mailhe so that the representative image set need only be processed in full once, and then those scores reused each time for an input image (see Vitsnudel, Col 5, Lines 29-35).
Regarding claim 6, Eusemann in view of Mailhe in view of Vitsnudel further teaches the method of claim 1,
wherein scoring comprises comparing the output to expert annotated data (see Eusemann, Paragraph [0045], “The output of the feature extraction at S260 may include any type of electronic data indicative of the extracted features. This data, along with the data generated at S250, may be used to generate a feature report at S270. The feature report may comprise a two-dimensional image and annotations describing the features identified/extracted at S250 and S260. In some embodiments, the data output by S250 and by S260 are not combined and are reviewed separately.”, the feature report is considered to comprise the scoring and annotated data).
The rationale of claim 1 has been applied herein.
Regarding claim 8, Eusemann in view of Mailhe in view of Vitsnudel further teaches the method of claim 1, further comprising:
performing a medical imaging procedure to acquire scan data (see Eusemann, Paragraph [0016], “detector 113 receives the radiation and produces a set of data (i.e., a raw image) for each projection angle”, a set of data is considered to be scan data);
reconstructing a first image from the scan data using the optimal reconstruction parameters (see Eusemann, Fig. 6 and Paragraph [0053], “In a particular example, the first set of imaging parameters used at S610 may be suited for a particular reconstruction technique in order to generate a three-dimensional image which is optimized for human identification of features.” a three-dimensional image which is optimized for human identification features is considered reconstruct a first image using optimal reconstruction parameters);
inputting the image into a computer aided diagnostic application configured for a clinical task (see Eusemann, Fig. 6 and Paragraph [0056], “The one or more three-dimensional images are generated at S650 using image reconstruction techniques which are intended to generate three-dimensional images suitable for automated feature extraction using computer-aided diagnosis systems” one or more three-dimensional images are suitable for automated feature extraction using computer-aided diagnosis systems which is considered to be input into the machine trained network);
and providing a diagnosis based on an output of the computer aided diagnostic application (see Eusemann, Fig. 6 and Paragraph [0059], “The output of S660 and S670 is used to generate a feature report at S680, which is correlated to a diagnosis at S69”).
The rationale of claim 1 has been applied herein.
Regarding claim 9, Eusemann in view of Mailhe in view of Vitsnudel further teaches the method of claim 8, further comprising:
reconstructing a second image from the scan data using a different set of reconstruction parameters (see Eusemann, Fig. 6 and Paragraph [0053], “At S620, it may be determined to acquire a second set of image data using a second set of imaging parameters which are suited for a particular reconstruction technique in order to generate a three-dimensional image which is optimized for a first automated feature extraction system”);
and displaying the second image for an operator (Eusemann Fig. 6 and Paragraph [0055], “A two-dimensional slice image of the image generated at S640 may then be displayed to an operator on terminal 130 and/or on another display of a separate computing system.”).
The rationale of claim 8 has been applied herein.
Claim(s) 2-3 is/are rejected under 35 U.S.C. 103 as being unpatentable over Eusemann et al, US 20190046129 in view of Mailhe et al, US 20200408864 in view of Vitsnudel et al, US 10210627 and in further view of Xia et al, US 20220130520.
Regarding claim 2, Eusemann in view of Mailhe in view of Vitsnudel does not expressively teach the method of claim 1,
wherein the reconstruction parameters comprise one or more of reconstruction algorithms, reconstruction kernels, pixel spacing, slice thickness and spacing, and beam hardening corrections.
However Xia in a similar invention in the same field of endeavor teaches the method of claim 1,
wherein the reconstruction parameters comprise one or more of reconstruction algorithms, reconstruction kernels, pixel spacing, slice thickness and spacing, and beam hardening corrections (see Xia, [0002], “as well as image reconstruction parameters such as reconstruction kernels, reconstruction algorithms, matrix size (e.g., 512, 1024), slice thickness, and patient-size dependent parameters”).
The combination of Eusemann, Mailhe, and Vitsnudel, and Xia are analogous art as they are all in the same field of endeavor of training a neural network and determining metrics of an input image. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, for reconstruction parameters to included reconstruction kernels, reconstruction algorithms, matrix size, slice thickness, and patient-size dependent parameters as taught in the method of Xia in the method of Eusemann in view of Mailhe in further view of Vitsnudel to maximize image quality while minimizing radiation exposure (see Xia, Abstract).
Regarding claim 3, Eusemann in view of Mailhe in view of Vitsnudel does not expressively teach the method of claim 1,
wherein reconstructing comprises reconstructing using a simulator.
However Xia in a similar invention in the same field of endeavor teaches the method of claim 1,
wherein reconstructing comprises reconstructing using a simulator (see Xia, [0042], “The system may include a computer simulation tool that can simulate CT images, which may be simulated two-dimensional (2D) CT images and/or simulated three-dimensional (3D) CT images, and corresponding dose map(s), from acquired scout scan information and scout scan conditions”).
The combination of Eusemann, Mailhe, Vitsnudel, and Xia are analogous art as they are in the same field of endeavor of medical image optimization. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, to include a computer simulation tool that can simulate CT images as taught in the method of Xia in view of Eusemann in view of Mailhe in view of Vitsnudel to maximize image quality while minimizing radiation exposure (see Xia, Abstract).
Claim(s) 4 and 7 is/are rejected under 35 U.S.C. 103 as being unpatentable over Eusemann et al, US 20190046129 in view of Mailhe et al, US 20200408864 in view of Vitsnudel et al, US 10210627 and in further view of Lu et al, US 20230326596.
Regarding claim 4, Eusemann in view of Mailhe in view of Vitsnudel does not expressively teach the method of claim 1,
wherein one or more combinations of the reconstruction parameters provide minimized regularization of the imaging data.
However Lu in a similar invention in the same field of endeavor teaches the method of claim 1,
wherein one or more combinations of the reconstruction parameters provide minimized regularization of the imaging data (see Lu, Paragraph [0128], “Moreover, a filter using a total-variation (TV) minimization regularization term can be used when an imaged region supports an assumption of uniformity over large areas demarked by sharp boundaries between uniform areas.”).
The combination of Eusemann, Mailhe, and Vitsnudel, and Lu are analogous art as they are all in the same field of endeavor of training a neural network and determining metrics of an input image. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, to use a filter with a minimization regularization term as taught in the method of Lu in the method of Eusemann in view of Mailhe in view of Vitsnudel to minimize the cost function (see Lu, Paragraph [0118]).
Regarding claim 7, Eusemann in view of Mailhe in view of Vitsnudel does not expressively teach the method of claim 1,
wherein the raw data comprises a CT sinogram
However Lu in a similar invention in the same field of endeavor teaches the method of claim 1,
wherein the raw data comprises a CT sinogram (see Lu, Fig 3. and Paragraph [0074], “. For example, as illustrated in FIG. 3, the projection data Y1 can be illustrated as a sinogram in which the channel direction of the X-ray detector 112 is set as a horizontal axis and the view (X-ray irradiation angle) is set as a vertical axis.”, projection data Y1 is considered to be raw data).
The combination of Eusemann, Mailhe, and Vitsnudel, and Lu are analogous art as they are all in the same field of endeavor of training a neural network and determining metrics of an input image. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, for the projection data to be illustrated as a sinogram as taught in the method of Eusemann in view of Mailhe in view of Vitsnudel to prepare training data used for machine learning and for noise reduction processing (see Lu, Paragraph [0002]).
Claim(s) 5 is/are rejected under 35 U.S.C. 103 as being unpatentable over Eusemann et al, US 20190046129 in view of Mailhe et al, US 20200408864 in view of Vitsnudel et al, US 10210627 and in further view of Mohr et al, US 20200118669.
Regarding claim 5, Eusemann in view of Mailhe in view of Vitsnudel does not expressively teach the method of claim 1,
wherein the given clinical task comprises coronary lumen segmentation or organ contouring.
However Mohr in a similar invention in the same field of endeavor teaches the method of claim 1,
wherein the given clinical task comprises coronary lumen segmentation or organ contouring (see Mohr, Paragraph [0094], “Similarly, a lumen segmentation parameter may be calculated based on data sets following the subtraction process, which has value(s) that depend on whether the determined lumen (e.g. blood vessel) segmentation is consistent with one or more predetermined geometrical, anatomical or other constraints.”).
The combination of Eusemann, Mailhe, and Vitsnudel, and Mohr are analogous art as they are both in the same field of endeavor of determining metrics from an input image. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, for the clinical task to comprise a lumen segmentation parameter as taught in the method of Mohr in the method of Eusemann in view of Mailhe in view of Vitsnudel to correct for motion between the cardiac phases (see Mohr, Abstract).
Claim(s) 10-11, and 15-16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Eusemann et al, US 20190046129 in view of Cheng et al, US 20180197317.
Regarding claim 10, Eusemann teaches a method
for generating an optimized machine trained network for clinical aided diagnostics of a given clinical task, the method comprising (see Eusemann, Paragraph [0039], “The one or more images may be generated using image processing techniques which are intended to generate images suitable for optimized and automated feature extraction using computer-aided diagnosis systems. For example, the one or more images may comprise three-dimensional images which are generated using image reconstruction techniques intended to generate three-dimensional images suitable for optimized and automated feature extraction using computer-aided diagnosis systems,” computer-aided diagnosis systems is considered to be clinical aided diagnostics given a clinical task):
reconstructing a plurality of sets of imaging data from a set of raw data, each set of the plurality of sets of imaging data using different combinations of reconstruction parameters (see Eusemann, Fig. 6 and Paragraph [0050], “process 600 utilizes multiple sets of image data, with each set of image data being acquired using different imaging parameters” and Paragraph [0054], “Two or more different sets of imaging parameters may be associated with a same reconstruction technique, and two or more different reconstruction techniques may be associated with a same set of imaging parameters”);
reconstructing a plurality of sets of imaging data from a set of raw data, each set of the plurality of sets of imaging data using different combinations of reconstruction parameters (see Eusemann, Fig. 6 and Paragraph [0050], “process 600 utilizes multiple sets of image data, with each set of image data being acquired using different imaging parameters” and Paragraph [0054], “Two or more different sets of imaging parameters may be associated with a same reconstruction technique, and two or more different reconstruction techniques may be associated with a same set of imaging parameters”);
Eusemann does not expressively teach
generating a plurality of different trained networks by machine training an original network, wherein each different trained network of the plurality of different trained networks uses a different set of the plurality of sets of imaging data;
comparing a performance of the plurality of different trained networks for the given clinical task;
and selecting the optimized machine trained network based the comparison.
However, Cheng in a similar invention in the same field of endeavor teaches
generating a plurality of different trained networks by machine training an original network, wherein each different trained network of the plurality of different trained networks uses a different set of the plurality of sets of imaging data (see Cheng, [0056], “it is possible to train a series of neural networks in this fashion, such as different models or networks for different stages of an iterative reconstruction process. For example, a first neural network may be trained by using early estimates (I.sub.m1, I.sub.m2, and I.sub.m3) as input and an estimate I.sub.N1 at iteration number N1 as the first target image where N1>m3. A second neural network may then be trained using I.sub.N1 as an input and an estimate I.sub.N2 at iteration N2 as the target image, where N2>N1. These sub-networks can be cascaded to form one deeper neural network and serve to pre-train and initialize a final, deeper network, so that the final network can achieve faster convergence and better performance,” different stages of an iterative reconstruction process are considered to be a plurality of sets of imaging data and a first neural network trained with early estimates and a second neural network trained with another set of estimates, where the sub-networks can be cascaded is considered to be generating a plurality of different trained networks by machine training an original network);
comparing a performance of the plurality of different trained networks for the given clinical task (Paragraph [0028], “the deep learning algorithms may process (either in a supervised or guided manner or in an unsupervised or unguided manner) the known or training data sets until the mathematical relationships between the initial data and desired output(s) are discerned and/or the mathematical relationships between the inputs and outputs of each layer are discerned and characterized. Similarly, separate validation data sets may be employed in which both input and desired target values are known, but only the initial values are supplied to the trained deep learning algorithms, with the outputs then being compared to the outputs of the deep learning algorithm to validate the prior training and/or to prevent over-training”, by comparing the mathematical relationships between the inputs and outputs with input and desired target value that is considered to be the comparing a performance; this process may be done on multiple neural networks);
and selecting the optimized machine trained network based the comparison (see Cheng, Paragraph [0047], “where I.sub.max corresponds to the optimal result where the cost function defined by the depicted curve is satisfied (here, maximized)”, I.sub.max is considered to be optimal reconstruction parameters in which would be compared to annotated data; Comparing the outputs was stated in argument above, the optimized machine trained network based on the comparison may be selected).
The combination of Eusemann and Cheng are analogous art as they are both in the same field of endeavor of training a neural network for optimizing medical images. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, to determine the optimal result where the cost function is satisfied; to acquire scan data (X-ray attenuation data) as taught in the method in the method of Cheng in the method of Eusemann to provide improved benefits, such as increased reconstruction efficiency or speed, while still achieving good image quality or allowing a low patient dose, achieving good image quality is important in Eusemann as generating multiple reconstructions and determining the optimal parameters is based on comparison to a reference image (see Cheng, Paragraph [0047]).
Regarding claim 11, Eusemann in view of Cheng further teaches the method of claim 10,
wherein the set of raw data comprises CT data (see Cheng, Paragraph [0032], “With this in mind, an example of an imaging system 110 (i.e., a scanner) is depicted in FIG. 2. In the depicted example, the imaging system 110 is a CT imaging system designed to acquire scan data (e.g., X-ray attenuation data).”, scan data (e.g., X-ray attenuation data) is considered to be raw data).
The rationale of claim 10 has been applied herein.
Regarding claim 15, Eusemann in view of Cheng further teaches the method of claim 10, further comprising:
performing a medical imaging procedure to acquire scan data (see Eusemann, Paragraph [0016], “detector 113 receives the radiation and produces a set of data (i.e., a raw image) for each projection angle”, a set of data is considered to be scan data);
reconstructing a first image from the scan data using the reconstruction parameters configured to provide minimized processing of the scan data (see Eusemann, Fig. 6 and Paragraph [0053], “In a particular example, the first set of imaging parameters used at S610 may be suited for a particular reconstruction technique in order to generate a three-dimensional image which is optimized for human identification of features.”);
inputting the image into the optimized machine trained network (see Eusemann, Fig. 2 and Paragraph [0049], “Similarly, the one or more images generated at S230 may be generated by inputting the one or more images to a trained network in order generate one or more images which are better-suited to automated feature extraction.”, Fig. 6 and Paragraph [0053], “In a particular example, the first set of imaging parameters used at S610 may be suited for a particular reconstruction technique in order to generate a three-dimensional image which is optimized for human identification of features.”, the one or more images generated at S230 are generated by inputting the images into a trained network and similarly for each set of imaging data at S650 are also input into a trained network);
and providing an output of the optimized machine trained network (see Eusemann, Fig. 6 and Paragraph [0059], “The output of S660 and S670 is used to generate a feature report at S680, which is correlated to a diagnosis at S69”).
The rationale of claim 10 has been applied herein.
Regarding claim 16, Eusemann in view of Cheng further teaches the method of claim 10,
wherein the network comprises a convolutional neural network (see Cheng, Paragraph [0051], “In certain embodiments, a convolutional neural network with or without POOL layers, fully convolutional network, recurrent neural network, Boltzmann machine, deep belief net, or the long short-term memory (LSTM) network, is employed”).
The rationale of claim 10 has been applied herein.
Claim(s) 12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Eusemann et al, US 20190046129 in view of Cheng et al, US 20180197317 and in further view of Xia et al, US 20220130520.
Regarding claim 12, Eusemann in view of Cheng does not expressively teach the method of claim 10,
wherein the reconstruction parameters comprise one or more of reconstruction algorithms, reconstruction kernels, pixel spacing, slice thickness and spacing, and beam hardening corrections.
However Xia in a similar invention in the same field of endeavor teaches the method of claim 10,
wherein the reconstruction parameters comprise one or more of reconstruction algorithms, reconstruction kernels, pixel spacing, slice thickness and spacing, and beam hardening corrections (see Xia, [0002], “as well as image reconstruction parameters such as reconstruction kernels, reconstruction algorithms, matrix size (e.g., 512, 1024), slice thickness, and patient-size dependent parameters.”).
The combination of Eusemann, Cheng, and Xia are analogous art as they are all in the same field of endeavor of training a neural network for optimizing medical images. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, for reconstruction parameters to included reconstruction kernels, reconstruction algorithms, matrix size, slice thickness, and patient-size dependent parameters as taught in the method of Xia in the method of Eusemann in view of Cheng to maximize image quality while minimizing radiation exposure (see Xia, Abstract).
Claim(s) 13 is/are rejected under 35 U.S.C. 103 as being unpatentable over Eusemann et al, US 20190046129 in view of Cheng et al, US 20180197317 and in further view of Lu et al, US 20230326596.
Regarding claim 13, Eusemann in view of Cheng does not expressively teach the method of claim 10,
wherein the different combinations of reconstruction parameters comprise reconstruction parameters configured to provide minimized processing of the raw data.
However Lu in a similar invention in the same field of endeavor teaches the method of claim 10,
wherein the different combinations of reconstruction parameters comprise reconstruction parameters configured to provide minimized processing of the raw data (see Lu, Paragraph [0128], “Moreover, a filter using a total-variation (TV) minimization regularization term can be used when an imaged region supports an assumption of uniformity over large areas demarked by sharp boundaries between uniform areas.”).
The combination of Eusemann, Cheng, and Lu are analogous art as they are all in the same field of endeavor of training a neural network for optimizing medical images. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, to use a filter with a minimization regularization term as taught in the method of Lu in the method of Eusemann in view of Cheng to minimize the cost function (see Lu, Paragraph [0118]).
Claim(s) 14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Cheng et al., U.S. Publication No. 2018/0197317 in further view of Mohr et al., U.S. Publication No. 2020/0118669.
Regarding claim 14, Eusemann in view of Cheng does not expressively teach the method of claim 10,
wherein the given clinical task comprises coronary lumen segmentation.
However Mohr in a similar invention in the same field of endeavor teaches the method of claim 10,
wherein the given clinical task comprises coronary lumen segmentation (see Mohr, Paragraph [0094], “Similarly, a lumen segmentation parameter may be calculated based on data sets following the subtraction process, which has value(s) that depend on whether the determined lumen (e.g. blood vessel) segmentation is consistent with one or more predetermined geometrical, anatomical or other constraints.”).
The combination of Eusemann, Cheng and Mohr are analogous art as they are all in the same field of endeavor of medical image processing. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, for the clinical task to comprise a lumen segmentation parameter as taught in the method of Mohr in the method of Eusemann in view of Cheng to correct for motion between the cardiac phases (see Mohr, Abstract).
Claim(s) 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over Eusemann et al, US 20190046129 in view of Dwivedi et al, US 11069098.
Regarding claim 19, Eusemann does not expressively teach the method of claim 17,
wherein the given clinical task comprises segmentation of an organ of a patient.
However Dwivedi in a similar invention in the same field of endeavor teaches the method of claim 17,
wherein the given clinical task comprises segmentation of an organ of a patient (see Dwivedi, Col 6, Lines 45-53, “for example, segmenting the organ or other anatomical feature of interest identified using the GUI 26 and adjusting the ROI to wholly contain this feature of interest, optionally with an additional margin. The counts data sub-set optimizer 30 then iteratively optimizes the data subset for reconstructing the ROI.”).
The combination of Eusemann and Dwivedi are analogous art as they are both in the same field of endeavor of medical image optimization. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention, for the clinical task to comprise of segmentation of an organ as taught in the in the method of Dwivedi in view of Eusemann to propose the segmented organ to the clinician as the ROI for enhanced image reconstruction (see Dwivedi, Col 5, Lines 41-45).
Conclusion
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DOMINIQUE JAMES whose telephone number is (703)756-1655. The examiner can normally be reached 9:00 am - 6:00 pm 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, Emily Terrell can be reached on (571)270-3717. 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.
/DOMINIQUE JAMES/Examiner, Art Unit 2666 /MING Y HON/Primary Examiner, Art Unit 2666
Prosecution Insights