DETAILED ACTION
Claims 1-10, 12-19 and 21-22 are pending in this application. Claims 1-2 and 12-13 have been amended, claims 11 and 20 have been canceled and claims 21-22 are newly added.
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 .
Response to Arguments
35 U.S.C. 112(b)
Applicant’s arguments (see Remarks filed 05/08/2026) regarding the previous rejections under 35 U.S.C. 112(b) have been fully considered by the examiner and are persuasive. Accordingly, the rejections made to claims 2 and 13 under 35 U.S.C. 112(b) have been withdrawn.
35 U.S.C. 102
Applicant’s arguments (see Remarks filed 05/08/2026) regarding the previous rejections under 35 U.S.C. 102(a)(1) have been fully considered by the examiner and are persuasive in view of the amendments to claims 1 and 13. In view of the amended change of scope to the claims, a new grounds of rejection is presented over Choi in view of Toporek as fully discussed below.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.+
Claims 1-4, 6-8, 12-15, 17-19 and 21-22 are rejected under 35 U.S.C. 103 as being unpatentable over Choi (20180330518 A1) in view of Toporek (US 20230100078 A1).
Regarding claim 1 Choi discloses; A method for analyzing ultrasound image data obtained from ultrasonic imaging, the method comprising (Choi, abstract, the system is for processing ultrasound image data and analyzing a region of interest):
obtaining, by an ultrasound probe, the ultrasound image data (Choi, [0024] the transducer probe is configured to acquire an image of a selected anatomical portion);
transforming, by a processor (Choi, [0026] the system has at least one processor), the ultrasound image data with at least one transform to generate at least one set of transformed data (Choi, [0037] the system performs a Fourier transform on the ultrasound data to generate FFT data (as shown in figure 4));
[ inputting the ultrasound image data and the at least one set of transformed data into a machine-learning model together, wherein the machine-learning model is implemented by the processor;]
implementing, by the processor (Choi, [0044] the machine learning functions may be executed by a processor), the machine-learning model with the ultrasound image data and the at least one set of transformed data (Choi, [0037] the CNN autoencoder unit may take the input spatial ultrasound image, and the FFT (transformed) image data as the inputs and generate feature maps using the input image data as shown in figure 4);
and identifying, by the processor (Choi, [0044] the machine learning functions may be executed by a processor), at least one feature in the ultrasound image data as determined by the machine-learning model (Choi, [0037] the CNN autoencoder unit may take the input spatial ultrasound image, and the FFT (transformed) image data as the inputs and generate feature maps (at least one feature using the input image data as shown in figure 4).
PNG
media_image1.png
282
510
media_image1.png
Greyscale
(Choi, Figure 4, Emphasis added)
Choi fails to disclose; inputting the ultrasound image data and the at least one set of transformed data into a machine-learning model together, wherein the machine-learning model is implemented by the processor;
However, in the same field of endeavor of image processing, Toporek teaches; inputting the ultrasound image data and the at least one set of transformed data into a machine-learning model together (Toporek, [0066] in training the machine learning model a pair of training data (Ius, Tus) are input into the model, where Ius is a 2D ultrasound image slice and Tus is a transformed ground truth ultrasound image), wherein the machine-learning model is implemented by the processor (Toporek, [0045] the processor is configured to execute a deep learning network);
PNG
media_image2.png
190
314
media_image2.png
Greyscale
(Toporek, [0066] emphasis added)
PNG
media_image3.png
134
508
media_image3.png
Greyscale
(Toporek, Figure 6)
The combination of Choi and Toporek would have been obvious to one of ordinary skill in the art prior to the effective filing date of the presently claimed invention. The motivation for use of the paired input of Toporek is that it allows both the transformation and the image data to be input as a training set where one is in the ground truth image allowing the model to be trained and validated. (Toporek, [0060]-[0070)
Regarding claim 2 the combination of Choi and Toporek teaches; The method of claim 1, further comprising limiting the ultrasound image data according to a region of interest (Choi, [0024]-[0025] the probe ultrasound signal ranges may be determined and adjusted based upon the region of interest being imaged).
Regarding claim 3 the combination of Choi and Toporek teaches; The method of claim 1, wherein the machine-learning model comprises a convolutional neural network (Choi, [0028] the system uses a Convolutional Neural Network).
Regarding claim 4 the combination of Choi and Toporek teaches; The method of claim 3, further comprising training the machine-learning model by:
inputting annotated ultrasound image data into the machine-learning model (Choi, [0037] the system may take b-mode images and feature maps as input (annotated image data)), wherein the ultrasound image data indicates the presence or absence of the at least one feature (Choi, [0073] the features identify a target of interest [0076] features are to be used in determining whether or not that the pixel belongs to the organ of interest, presences of target or not is determined by the features);
transforming the ultrasound image data using at least one transform to generate transformed data (Choi, [0037] the system performs a Fourier transform on the ultrasound data to generate FFT data (as shown in figure 4));
and inputting the transformed data into the machine-learning model (Choi, [0037] the CNN autoencoder unit may take the input spatial ultrasound image, and the FFT (transformed) image data as the inputs and generate feature maps using the input image data as shown in figure 4).
Regarding claim 6 the combination of Choi and Toporek teaches; The method of claim 1, wherein the ultrasound image data comprises spatial B-mode image data (Choi, [0016] the image is a B-mode image, [0029] the b-mode image is generated based on return echoes, and is 2D making it a spatial b-mode image).
Regarding claim 7 the combination of Choi and Toporek teaches; The method of claim 1, wherein the at least one set of transformed data comprises at least one of Fourier transformed data, slant transformed data, or Hadamard transformed data (Choi, [0037] the CNN autoencoder unit may take the input spatial ultrasound image, and the FFT (Fourier transform) (transformed) image data as the inputs and generate feature maps using the input image data as shown in figure 4).
Regarding claim 8 the combination of Choi and Toporek teaches; The method of claim 7, wherein the at least one set of transformed data comprises only one of Fourier transformed data, slant transformed data, or Hadamard transformed data (Choi, [0037] the CNN autoencoder unit may take the input spatial ultrasound image, and the FFT (Fourier transform) (transformed) image data as the inputs and generate feature maps using the input image data as shown in figure 4, where only an FFT (Fourier transform) is done ).
Regarding claim 12 the combination of Choi and Toporek teaches; A system for analyzing ultrasound image data obtained from ultrasonic imaging, the system comprising:
an ultrasound probe configured to obtain the ultrasound image data (Choi, [0024] the transducer probe is configured to acquire an image of a selected anatomical portion);
and a processor (Choi, [0026] the system has at least one processor) configured to transform the ultrasound image data with at least one transform to generate at least one set of transformed data (Choi, [0037] the system performs a Fourier transform on the ultrasound data to generate FFT data (as shown in figure 4)),
input the ultrasound image data and the at least one set of transformed data into a machine-learning model (Choi, [0037] the CNN autoencoder unit may take the input spatial ultrasound image, and the FFT (transformed) image data as the inputs and generate feature maps using the input image data as shown in figure 4), wherein the machine-learning model is implemented by the processor(Choi, [0044] the machine learning functions may be executed by a processor),
implement the machine-learning model with the ultrasound image data and the at least one set of transformed data (Choi, [0037] the CNN autoencoder unit may take the input spatial ultrasound image, and the FFT (transformed) image data as the inputs and generate feature maps using the input image data as shown in figure 4),
and identify at least one feature in the ultrasound image data as determined by the machine-learning model together (Toporek, [0066] in training the machine learning model a pair of training data (Ius, Tus) are input into the model, where Ius is a 2D ultrasound image slice and Tus is a transformed ground truth ultrasound image).
The combination of Choi and Toporek would have been obvious to one of ordinary skill in the art prior to the effective filing date of the presently claimed invention. The motivation for use of the paired input of Toporek is that it allows both the transformation and the image data to be input as a training set where one is in the ground truth image allowing the model to be trained and validated. (Toporek, [0060]-[0070)
Regarding claim 13 the combination of Choi and Toporek teaches; The system of claim 12, wherein the processor is further configured limit the ultrasound image data according to a region of interest (Choi, [0024]-[0025] the probe ultrasound signal ranges may be determined and adjusted based upon the region of interest being imaged).
Regarding claim 14 the combination of Choi and Toporek teaches; The system of claim 12, wherein the machine-learning model comprises a convolutional neural network(Choi, [0028] the system uses a Convolutional Neural Network).
Regarding claim 15 the combination of Choi and Toporek teaches; The method of claim 14, wherein the processor is further configured to train the machine-learning model by inputting annotated ultrasound image data into the machine-learning model (Choi, [0037] the system may take b-mode images and feature maps as input (annotated image data)), wherein the ultrasound image data indicates the presence or absence of the at least one feature (Choi, [0073] the features identify a target of interest [0076] features are to be used in determining whether or not that the pixel belongs to the organ of interest, presences of target or not is determined by the features), transforming the ultrasound image data using at least one transform to generate transformed data (Choi, [0037] the system performs a Fourier transform on the ultrasound data to generate FFT data (as shown in figure 4)), and inputting the transformed data into the machine-learning model (Choi, [0037] the CNN autoencoder unit may take the input spatial ultrasound image, and the FFT (transformed) image data as the inputs and generate feature maps using the input image data as shown in figure 4).
Regarding claim 17 the combination of Choi and Toporek teaches; The system of claim 12, wherein the ultrasound image data comprises spatial B-mode image data (Choi, [0016] the image is a B-mode image, [0029] the b-mode image is generated based on return echoes, and is 2D making it a spatial b-mode image).
Regarding claim 18 the combination of Choi and Toporek teaches; The system of claim 12, wherein the at least one set of transformed data comprises at least one of Fourier transformed data, slant transformed data, or Hadamard transformed data (Choi, [0037] the CNN autoencoder unit may take the input spatial ultrasound image, and the FFT (Fourier transform) (transformed) image data as the inputs and generate feature maps using the input image data as shown in figure 4, where only an FFT (Fourier transform) is done ).
Regarding claim 19 the combination of Choi and Toporek teaches; The method of claim 12, wherein the at least one set of transformed data comprises only one of Fourier transformed data, slant transformed data, or Hadamard transformed data (Choi, [0037] the CNN autoencoder unit may take the input spatial ultrasound image, and the FFT (Fourier transform) (transformed) image data as the inputs and generate feature maps using the input image data as shown in figure 4, where only an FFT (Fourier transform) is done).
Regarding claim 21 the combination of Choi and Toporek teaches; The method of claim 1, further comprising receiving at, at least one filter in the machine-learning model, the ultrasound image data and the at least one set of transformed data together (Toporek, [0061] and [0067] the machine learning model has filters in the layers that are adjusted based on received data, this filter adjustment/filter weighting is passed through the model based on the training data).
The combination of Choi and Toporek would have been obvious to one of ordinary skill in the art prior to the effective filing date of the presently claimed invention. The motivation for use of the filters of Toporek is that this allows for extraction of features of the training data and customization of the filtering based on the input data. (Toporek, [0060]-[0070)
Regarding claim 22 the combination of Choi and Toporek teaches; The system of claim 12, wherein the processor is further configured to receive at, at least one filter in the machine-learning model, the ultrasound image data and the at least one set of transformed data together (Toporek, [0061] and [0067] the machine learning model has filters in the layers that are adjusted based on received data, this filter adjustment/filter weighting is passed through the model based on the training data).
The combination of Choi and Toporek would have been obvious to one of ordinary skill in the art prior to the effective filing date of the presently claimed invention. The motivation for use of the filters of Toporek is that this allows for extraction of features of the training data and customization of the filtering based on the input data. (Toporek, [0060]-[0070)
2. Claims 5, and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Choi (20180330518 A1) in view of Toporek (US 20230100078 A1) and Vega (20240273726 A1).
Regarding claim 5 the combination of Choi and Toporek does not disclose; The method of claim 4, further comprising updating the machine-learning model to reduce a loss function as the machine-learning model receives additional ultrasound image data indicating the presence or absence of the at least one feature.
However, in the same field of endeavor, Vega teaches; further comprising updating the machine-learning model to reduce a loss function as the machine-learning model receives additional ultrasound image data indicating the presence or absence of the at least one feature (Vega, [0104] the network is trained to detect features, and a loss function is computed during training, then the loss function is used to update the model weights to reduce loss).
The combination of Choi, Toporek and Vega would have been obvious to one of ordinary skill in the art prior to the effective filing date of the presently claimed invention. Choi and Vega both teach methods of training a machine learning model to identify features of b-mode ultrasound images, however Choi does not teach updating the model to reduce loss. The motivation to add the function to reduce loss in this way is advantageous to help increase the accuracy of the predictions (Vega, [0103]-[0105]).
Regarding claim 16 the combination of Choi, Toporek, and Vega teaches; The method of claim 15, wherein the processor is further configured to implement the machine-learning model to reduce a loss function as the machine-learning model receives additional ultrasound image data indicating the presence or absence of the at least one feature(Vega, [0104] the network is trained to detect features, and a loss function is computed during training, then the loss function is used to update the model weights to reduce loss).
The combination of Choi, Toporek, and Vega would have been obvious to one of ordinary skill in the art prior to the effective filing date of the presently claimed invention. Choi and Vega both teach methods of training a machine learning model to identify features of b-mode ultrasound images, however Choi does not teach updating the model to reduce loss. The motivation to add the function to reduce loss in this way is advantageous to help increase the accuracy of the predictions (Vega, [0103]-[0105]).
3. Claims 9 and 10 rejected under 35 U.S.C. 103 as being unpatentable over Choi (20180330518 A1) in view of Toporek (US 20230100078 A1) and in further view of Powell (EP 0126148 B1).
Regarding claim 9 the combination of Choi and Toporek does not teach; The method of claim 7, wherein the at least one set of transformed data comprises only two of Fourier transformed data, slant transformed data, or Hadamard transformed data.
However, in the same field of endeavor of image processing, Powell teaches; wherein the at least one set of transformed data comprises only two of Fourier transformed data, slant transformed data, or Hadamard transformed data (Powell, Column 17, lines 30-58, the system performs 2 stage denoising, where the first stage is a slant transform and the second is a Walsh-Hadamard transform).
PNG
media_image4.png
340
278
media_image4.png
Greyscale
(Powell, Column 17)
The combination of Choi, Toporek and Powell would have been obvious to one of ordinary skill in the art prior to the filing date of the presently claimed invention. Choi teaches a method of processing an ultrasound image using machine learning, but does not teach the use of multiple different transforms to generate transformed data. Powell teaches this deficiency, teaching a method of using multi-stage image transforms to denoise images. The motivation for the addition of the use of multiple types of data transforms as taught by Powell is that using multiple image transforms over multiple stages can help reduce image noise visibly and minimize loss. (Powell, Abstract)
Regarding claim 10 the combination of Choi, Toporek and Powell teaches; The method of claim 7, wherein the at least one set of transformed data comprises Fourier transformed data, slant transformed data, and Hadamard transformed data (Powell figures 1 and 2 show that the system is a multi-stage transform method for denoising images, where it feeds image data into multiple direct transform networks (3 transforms), Column 8 line 64 through column 9 line 20 notes that these “direct transform blocks” may be any transform including a Fourier, Slant or Walsh-Hadamard.
PNG
media_image5.png
210
292
media_image5.png
Greyscale
(Powell, column 9)
PNG
media_image6.png
330
580
media_image6.png
Greyscale
(Powell, figure 2, emphasis added)
The combination of Choi, Toporek and Powell would have been obvious to one of ordinary skill in the art prior to the filing date of the presently claimed invention. Choi teaches a method of processing an ultrasound image using machine learning, but does not teach the use of multiple different transforms to generate transformed data. Powell teaches this deficiency, teaching a method of using multi-stage image transforms to denoise images. The motivation for the addition of the use of multiple types of data transforms as taught by Powell is that using multiple image transforms over multiple stages can help reduce image noise visibly and minimize loss. (Powell, Abstract)
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). 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.
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure. For a listing of analogous prior art as cited by the examiner, please see the attached PTO 892- Notice of References Cited page.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to JORDAN M ELLIOTT whose telephone number is (703)756-5463. The examiner can normally be reached M-F 8AM-5PM ET.
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 at (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.
/J.M.E./Examiner, Art Unit 2666
/EMILY C TERRELL/Supervisory Patent Examiner, Art Unit 2666
Prosecution Insights