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 .
Applicant’s response to the Non-final Office Action dated 03/25/2026, filed with the office on 06/25/2026, has been entered and made of record.
Status of Claims
Claims 1-7, 9-11, 13-17 and 19-24 are pending. Claims 1, 5, 16 and 21 are amended. Claims 8, 12 and 18 were previously cancelled. Claims 16, 17, 19 and 20 are allowable.
Response to Arguments
Applicant’s amendment of the Claims, which has altered the scope of the claims of the instant application, has necessitated the new ground(s) of rejection presented in this office action with respect to claims of the instant application. Accordingly, in response to Applicant’s arguments that are merely directed to the amended portion of the claims, new analyses have been presented below, which make Applicant’s arguments moot.
Consequently, THIS ACTION IS MADE FINAL.
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.
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.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-7, 9-11 and 13 are rejected under 35 U.S.C. 103 as being unpatentable over Bai et al. (US 2020/0334870 A1), in view of Ye et al. (US 2014/0334701 A1), in further view of Chan et al. (US 2019/0365341 A1) and still in further view of Abou Shousha et al. (US 10,468,142 B1).
Regarding claim 1, Bai teaches, A method for a Positron Emission Tomography (PET) system, the method comprising: performing a first reconstruction of a first image volume (Bai, ¶0002: “image reconstruction is commonly used in reconstructing emission imaging data such as positron emission tomography (PET)”) using raw image data of the PET system (Bai, ¶0033: “the system 10 includes an image acquisition device 12. In one example, the image acquisition device 12 can comprise a PET imaging device”) without specifying a noise regularization parameter; (Bai, ¶0021: “for this optimization to involve a single reconstruction using some default value (e.g. without any regularization”) (Bai, ¶0030: “allowing automatic parameter tuning to obtain the baseline β parameter for each patient”) based on preference data of the user, (Bai, ¶0037: “weighting parameter is selected based on the received user input indicative of the noise reduction goal”) wherein the (Bai, ¶0006: “selecting the weighting parameter based on a received user input indicative of the noise reduction goal for the reconstructed image”) performing a second reconstruction of a second image volume using the raw image data of the PET system and the automatically customized, predicted noise regularization parameter setting, (Bai, ¶0044: “generate a reconstructed image by applying the regularized image reconstruction including the edge-preserving regularization or penalty with the determined weighting and edge sensitivity parameters”) the second image volume having a higher image quality than the first image volume; (Bai, ¶0046: “The signal to noise ratio was improved from 7.84 in the conventionally reconstructed image 34 to 16.88 in the regularized reconstruction image”) and displaying the second image volume on a display device of the PET system. (Bai, ¶0042: “display device 24 to display a plurality of sample reconstructed images generated by reconstructing a sample imaging data set by applying the regularized image reconstruction including the edge-preserving regularization or penalty with different weighting and edge sensitivity parameters to the sample imaging data set”). However, Bai does not explicitly teach extracting a plurality of patches from the first reconstruction of the first image volume; using a first trained model to automatically predict a noise regularization parameter setting based on the patches extracted from the reconstruction of the first image volume, wherein the noise regularization parameter setting establishes how close neighboring voxels in the image volume have to be in terms of absolute value; inputting the predicted noise regularization parameter setting output by the first trained model into a second trained model.
In an analogous field of endeavor, Ye teaches, extracting a plurality of patches from the first reconstruction of the first image volume; (Ye, ¶0030: “in the initial reconstruction image… extracting the predetermined number of similar patches”) using a first trained model to automatically predict a noise regularization parameter setting (Ye, ¶0099: “An algorithm that updates a reconstruction image based on a patch-based low rank regularization image may be employed to reduce noise in the reconstruction image”) based on the patches extracted from the reconstruction of the first image volume, (Ye, ¶0099: “regularization image may be generated based on patches obtained from the reconstruction image”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai using the teachings of Ye to introduce extracting patches from a reconstructed image. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of automatically reducing noise in a reconstructed PET image based on the extracted patches. Therefore, it would have been obvious to combine the analogous arts Bai and Ye to obtain the above-described limitations in claim 1. However, the combination of Bai and Ye does not explicitly teach, wherein the noise regularization parameter setting establishes how close neighboring voxels in the image volume have to be in terms of absolute value; inputting the predicted noise regularization parameter setting output by the first trained model into a second trained model.
In another analogous field of endeavor, Chan teaches, wherein the noise regularization parameter setting establishes how close neighboring voxels in the image volume have to be in terms of absolute value; (Chan, ¶0035: “cost function that is optimized to iteratively reconstruct an image of the activity level (e.g., tracer density) within the respective voxels (i.e., volume pixels) of the reconstructed image”; how close neighboring voxels are located in image is interpreted as the density of the voxels in image).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai in view of Ye using the teachings of Chan to introduce a neural network. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of automatically adjusting a noise regularization parameter to reconstruct a PET image. Therefore, it would have been obvious to combine the analogous arts Bai and Chan to obtain the above-described limitations in claim 1. However, the combination of Bai, Ye and Chan does not explicitly teach, inputting the predicted noise regularization parameter setting output by the first trained model into a second trained model.
In yet another analogous field of endeavor, Abou Shousha teaches, inputting the predicted noise regularization parameter setting output (Abou Shousha, col. 22, lines 50-51: “output of a first CNN may be provided as an input”) by the first trained model (Abou Shousha, col. 19, lines 16-17: “AI model 12 has been trained”) into a second trained model (Abou Shousha, col. 19, lines 27-29: “The score, input data, or both may then be processed through one or more second submodels trained to generate second sets of scores corresponding to a prediction”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai in view of Ye, and in further view of Chan using the teachings of Abou Shousha to introduce inputting an output of a trained model to a second trained model. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of customizing a prediction from a neural network based on user preference. Therefore, it would have been obvious to combine the analogous arts Bai, Ye, Chan and Abou Shousha to obtain the invention in claim 1.
Regarding claim 2, Bai in view of Ye, in further view of Chan and in further view of Abou Shousha teaches, The method of claim 1, further comprising reconstructing the first image volume using fast ordered subsets expectation maximization (OSEM) reconstruction. (Bai, ¶0046: “ordered subset expectation maximization (OSEM) reconstruction of the image 34”).
Regarding claim 3, Bai in view of Ye, in further view of Chan and in further view of Abou Shousha teaches, The method of claim 1, wherein the first trained model is a convolutional neural network (CNN) trained (Chan, ¶0041: “a range of noise levels used in training the DL-CNN network”) on a plurality of patches extracted from image volumes (Chan, ¶0002: “a feature oriented training strategy… voxels corresponding to lesion and/or regions of interest”) having different noise levels, (Chan, ¶0033: “convolutional neural networks (CNN) to obtain consistently high-quality images from PET data with a large degree of variability (e.g., different noise levels”) each patch of the plurality of patches labeled with a target noise regularization parameter setting. (Chan, ¶0097: “ground truth labels (i.e., the high quality image 153”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai in view of Ye, in further view of Chan and in further view of Abou Shousha using the additional teachings of Chan to introduce training patches with different noise levels. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of optimized image reconstruction for different noise levels. Therefore, it would have been obvious to combine the analogous arts Bai, Ye, Chan and Abou Shousha to obtain the invention in claim 3.
Regarding claim 4, Bai in view of Ye, in further view of Chan and in further view of Abou Shousha teaches, The method of claim 3, wherein the extracted patches are extracted from portions of the first image volume located at different bed positions of the PET system, (Chan, ¶0072: “weight maps have higher value in the lesions and lower value in the background, and they can be uniform for the patches that do not contain lesions”) and using the first trained model to predict the noise regularization parameter setting further comprises: using the first trained model to predict a bed noise regularization parameter (Chan, ¶0055: “train the DL-CNN network 162 using reconstructed images have a variety of noise levels, making the DL-CNN network robust to variations in the noise level”) for each bed position of the different bed positions; (Chan, ¶0072: “higher value in the lesions and lower value in the background, and they can be uniform for the patches that do not contain lesions”; bed positions are interpreted as regions with lesions, no lesions and background) and one of: averaging the predicted bed noise regularization parameter settings to generate the second noise regularization parameter setting; (Chan, ¶0069: “denoised 3-D volumes are then combined (e.g., averaged) to obtain the final result—the high-quality image 253”) and using the bed noise regularization parameters for each bed position in the second reconstruction. (Chan, ¶0033: “produce output images having a uniform image quality, even when the input images exhibit large variations in their image quality and statistical properties (e.g., by having different noise levels”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai in view of Ye, in further view of Chan and in further view of Abou Shousha using the additional teachings of Chan to introduce averaging denoised patch volumes. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of reconstructing uniformly denoised image. Therefore, it would have been obvious to combine the analogous arts Bai, Ye, Chan and Abou Shousha to obtain the invention in claim 4.
Regarding claim 5, Bai in view of Ye, in further view of Chan and in further view of Abou Shousha teaches, The method of claim 1, wherein the preference data of the user is based on ratings of the user of images having different noise levels and noise regularization parameters, that are displayed to the user in pairings or groupings. (Bai, ¶0042: “receive a user selection via the user input device 22 of one of the displayed plurality of sample reconstructed images”).
Regarding claim 6, Bai in view of Ye, in further view of Chan and in further view of Abou Shousha teaches, The method of claim 1, wherein: for a first operator of a PET imaging system, the second reconstruction of the second image volume is performed with a first noise regularization value customized (Bai, ¶0021: “liver noise information from the single reconstruction) is input to a lookup table or an empirical calibration curve to select β generated from phantom data”) for the first operator, (Bai, ¶0021: “optimize both the weighting parameter β and the edge preservation parameter γ for… a particular patient”) and the second image volume is displayed on the display device (Bai, ¶0042: “display device 24 to display a plurality of sample reconstructed images generated by reconstructing a sample imaging data set by applying the regularized image reconstruction”) with a first CNR; (Bai, ¶0019: “increasing β leads to more noise reduction, but results in more image blurring, lower contrast, reduced quantitation, e.g., tumor SUV. Increasing γ leads to sharper image, higher contrast”) and for a second operator of the PET imaging system, the second reconstruction of the second image volume is performed with a second noise regularization value customized (Bai, ¶0021: “liver noise information from the single reconstruction) is input to a lookup table or an empirical calibration curve to select β generated from phantom data”) for the second operator, (Bai, ¶0041: “the noise may depend on the patient's girth… values for β depending on both the noise goal and patient girth”) and the second image volume is displayed on the display device (Bai, ¶0042: “display device 24 to display a plurality of sample reconstructed images generated by reconstructing a sample imaging data set by applying the regularized image reconstruction”) with a second CNR, (Bai, ¶0019: “increasing β leads to more noise reduction, but results in more image blurring, lower contrast, reduced quantitation, e.g., tumor SUV. Increasing γ leads to sharper image, higher contrast”) the second CNR different from the first CNR. (Bai, ¶0046: “using the determined parameters β and γ, achieve a doubled signal to noise ratio in liver in regularized image reconstruction as compared to the conventional list-mode ordered subset expectation maximization (OSEM) reconstruction”).
Regarding claim 7, Bai teaches, A hybrid recommendation system (RS) for recommending a parameter setting (Bai, ¶0019: “Obtaining the optimal setting of the β and γ values”) for acquiring and/or reconstructing an image via an imaging system, (Bai, ¶0006: “an imaging system includes an image acquisition device configured to acquire imaging data. At least one electronic processor is programmed to: determine a weighting parameter of an edge-preserving regularization or penalty of a regularized image reconstruction”) the hybrid RS comprising: (Bai, ¶0037: “weighting parameter is selected based on the received user input indicative of the noise reduction goal for the reconstructed image”). However, Bai does not explicitly teach, a first model trained to automatically predict a parameter setting based on a preliminary reconstructed image-, wherein the parameter setting comprises a noise regularization parameter setting that establishes how close neighboring voxels in the preliminary reconstructed image have to be in terms of absolute value; and a second model trained to automatically customize the predicted parameter setting output by the first model.
In an analogous field of endeavor, Chan teaches, a first model trained (Chan, ¶0082: “a DL-CNN network trained using different weights”) to automatically predict a parameter setting based on a preliminary reconstructed image-, (Chan, ¶0082: “denoising an OS-EM reconstructed image by applying DL-CNN networks using either a uniform weighting or different weights determined by weight maps”) wherein the parameter setting comprises a noise regularization parameter setting that establishes how close neighboring voxels in the preliminary reconstructed image have to be in terms of absolute value; (Chan, ¶0035: “In a regularization approach, a user defined penalty function is incorporated into an objective/cost function that is optimized to iteratively reconstruct an image of the activity level (e.g., trader density) within the respective voxels (i.e., volume pixels) of the reconstructed image. The user defined penalty function can, e.g., encourage local smoothness and thus supress noise”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai using the teachings of Chan to introduce a neural network. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of automatically adjusting noise compensation parameters to reconstruct a PET image. Therefore, it would have been obvious to combine the analogous arts Bai and Chan to obtain the above-described limitations in claim 7. However, the combination of Bai and Chan does not explicitly teach, a second model trained to automatically customize the predicted parameter setting output by the first model.
In another analogous field of endeavor, Abou Shousha teaches, a second model trained to automatically customize the predicted parameter setting (Abou Shousha, col. 19, lines 27-29: “The score, input data, or both may then be processed through one or more second submodels trained to generate second sets of scores corresponding to a prediction”) output by the first model (Abou Shousha, col. 22, lines 50-51: “output of a first CNN may be provided as an input”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai in view of Chan using the teachings of Abou Shousha to introduce inputting an output of a trained model to a second trained model. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of customizing a prediction from a neural network based on user preference. Therefore, it would have been obvious to combine the analogous arts Bai, Ye, Chan and Abou Shousha to obtain the invention in claim 7.
Regarding claim 9, Bai in view of Ye, in further view of Chan and in further view of Abou Shousha teaches, The hybrid RS of claim 7, wherein the imaging system includes a Positron Emission Tomography (PET) system, (Bai, ¶0002: “reconstructing emission imaging data such as positron emission tomography (PET) or single photon emission computed tomography (SPECT) imaging”) and the parameter setting is an acquisition time per bed used for acquiring the image. (Chan, ¶0081: “different acquisition durations (e.g., 1, 2, 3 and 4 min) were simulated to generate different noise levels. The high quality target images were simulated with 10-min acquisition”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai in view of Ye, in further view of Chan and in further view of Abou Shousha using the additional teachings of Chan to introduce different acquisition times. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of determining optimal setting for desired image quality. Therefore, it would have been obvious to combine the analogous arts Bai, Ye, Chan and Abou Shousha to obtain the invention in claim 9.
Regarding claim 10, Bai in view of Ye, in further view of Chan and in further view of Abou Shousha teaches, The hybrid RS of claim 7, wherein the first model is a convolutional neural network (CNN) trained on a plurality of patches extracted from image volumes including different noise levels, (Chan, ¶0033: “convolutional neural networks (CNN) to obtain consistently high-quality images from PET data with a large degree of variability (e.g., different noise levels”) each patch of the plurality of patches labeled with a target noise regularization parameter setting. (Chan, ¶0097: “ground truth labels (i.e., the high quality image 153”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai in view of Ye, in further view of Chan and in further view of Abou Shousha using the additional teachings of Chan to introduce training patches with different noise levels. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of optimized image reconstruction for different noise levels. Therefore, it would have been obvious to combine the analogous arts Bai, Ye, Chan and Abou Shousha to obtain the invention in claim 10.
Regarding claim 11, Bai in view of Ye, in further view of Chan and in further view of Abou Shousha teaches, The hybrid RS of claim 10, wherein the image volumes including different noise levels are generated by performing a plurality of scans of an anatomy of a subject, (Bai, ¶0030: “whole body scans are performed, the disclosed tuning approach can be fully automatic”) each scan having a different noise regularization parameter setting. (Bai, ¶0031: “tune the penalty lower when a voxel has an SUV value 5 or above”).
Regarding claim 13, Bai in view of Ye, in further view of Chan and in further view of Abou Shousha teaches, The hybrid RS of claim 7, wherein the second model maps the predicted parameter setting to a customized parameter setting for the user, (Bai, ¶0021: “optimize both the weighting parameter β and the edge preservation parameter γ for… a particular patient”) based on preference data of the user collected from user ratings of images reconstructed with a range of noise levels. (Bai, ¶0042: “receive a user selection via the user input device 22 of one of the displayed plurality of sample reconstructed images”).
Claims 14 and 15 are rejected under 35 U.S.C. 103 as being unpatentable over Bai et al. (US 2020/0334870 A1), in view of Ye et al. (US 2014/0334701 A1) in further view of Chan et al. (US 2019/0365341 A1), still in further view of Abou Shousha et al. (US 10,468,142 B1) and yet in further view of Lang et al. (US 6,308,175 B1).
Regarding claim 14, Bai in view of Ye, in further view of Chan and still in further view of Abou Shousha teaches, The hybrid RS of claim 13. However, the combination of Bai, Ye, Chan and Abou Shousha does not explicitly teach wherein the preference data is filtered using content and collaborative filtering.
In an analogous field of endeavor, Lang teaches, wherein the preference data is filtered using content and collaborative filtering. (Lang, col 1, lines 44-47: “an advanced collaborative/content-based information filter system is employed to provide superior filtering in the process of finding and rating informons which match a user's query”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai in view of Ye, in further view of Chan and still in further view of Abou Shousha using the teachings of Lang to introduce content and collaborative filtering. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of selecting parameter setting based on the preference of single user as well as a group of users. Therefore, it would have been obvious to combine the analogous arts Bai, Ye, Chan, Abou Shousha and Lang to obtain the invention in claim 14.
Regarding claim 15, Bai in view of Ye, in further view of Chan and in further view of Abou Shousha teaches, The hybrid RS of claim 13. However, the combination of Bai, Ye, Chan and Abou Shousha does not explicitly teach wherein the preference data is used to classify the user to a group of users with similar noise preferences.
In an analogous field of endeavor, Lang teaches, wherein the preference data is used to classify the user to a group of users with similar noise preferences. (Lang, col. 8, lines 61-65: “two users may provide similar evaluative feedback for the informons which they both consider. As a result of adaptive adjustment to feedback data, the concept data profiles of the two users may become similar”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai in view of Ye, in further view of Chan and in further view of Abou Shousha using the teachings of Lang to introduce user profile similarity. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of classifying similar users in a same group. Therefore, it would have been obvious to combine the analogous arts Bai, Ye, Chan, Abou Shousha and Lang to obtain the invention in claim 15.
Claim 21 is rejected under 35 U.S.C. 103 as being unpatentable over Bai et al. (US 2020/0334870 A1), in view of Ye et al. (US 2014/0334701 A1), in further view of Chan et al. (US 2019/0365341 A1), still in further view of Abou Shousha et al. (US 10,468,142 B1) and yet in further view of Yang et al. (US 2020/0105032 A1).
Regarding claim 21, Bai in view of Ye, in further view of Chan and in further view of Abou Shousha teaches, The method of claim 1, wherein the first trained model is used to wherein the first trained model assigns a beta value to each bed position of the different bed positions, (Bai, ¶0030: “the process repeats with increased regularization (by increasing β slightly), until the noise level is met, allowing automatic parameter tuning to obtain the baseline β parameter”) and the method further comprises independently inputting the predicted noise regularization parameter settings for each of the different bed positions into the second trained model (Abou Shousha, col. 40, lines 7-12: “AI model 12 outputs the probability of the input patch… repeating this for all pixels, a probability map of all pixels belonging to each category may be constructed”) to be customized by the second trained model. (Abou Shousha, col. 19, lines 27-29: “The score, input data, or both may then be processed through one or more second submodels trained to generate second sets of scores corresponding to a prediction”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai in view of Ye, in further view of Chan and in further view of Abou Shousha using the additional teachings of Abou Shousha to introduce inputting different patches representing different category. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of optimizing different parameters corresponding to different positions of the scan. Therefore, it would have been obvious to combine the analogous arts Bai, Ye, Chan and Abou Shousha to obtain the above-described limitations in claim 21. However, the combination of Bai, Ye, Chan and Abou Shousha does not explicitly teach, predict distinct noise regularization parameter settings for different portions of the first image volume located at different bed positions of the PET system, wherein the first trained model assigns a beta value to each bed position of the different bed positions, .
In an analogous field of endeavor, Yang teaches, predict distinct noise regularization parameter settings for different portions of the first image volume located at different bed positions of the PET system, (Yang, ¶0041: “tradeoff between resolution and noise is addressed through the use of a spatially varying map for the regularization parameter β({right arrow over (r)}), thereby suppressing noise within low activity regions while preserving image features such as fine structures”) wherein the first trained model assigns a beta value to each bed position of the different bed positions, (Yang, ¶0041: “tradeoff between resolution and noise is addressed through the use of a spatially varying map for the regularization parameter β({right arrow over (r)}), thereby suppressing noise within low activity regions while preserving image features such as fine structures”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai in view of Ye, in further view of Chan and in further view of Abou Shousha using the teachings of Yang to introduce varying noise suppression. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of optimized denoising in different positions of the PET scan. Therefore, it would have been obvious to combine the analogous arts Bai, Ye, Chan, Abou Shousha and Yang to obtain the invention in claim 21.
Claims 22-24 are rejected under 35 U.S.C. 103 as being unpatentable over Bai et al. (US 2020/0334870 A1), in view of Ye et al. (US 2014/0334701 A1), in further view of Chan et al. (US 2019/0365341 A1), still in further view of Abou Shousha et al. (US 10,468,142 B1) and yet in further view of Chan et al. (US 20190236763 A1, herein referred to as Chan’763).
Regarding claim 22, Bai in view of Ye, in further view of Chan and in further view of Abou Shousha teaches, The method of claim 1, wherein the predicted noise regularization parameter setting is output as a recommended beta value (Bai, ¶0021: “optimize both the weighting parameter β and the edge preservation parameter γ for… a particular patient”). However, the combination of Bai, Ye, Chan and Abou Shousha does not explicitly teach, wherein the raw image data of the first image volume is acquired at different bed positions of the PET system, and wherein different beta values are output for different portions of the raw image data of the first image volume acquired at the different bed positions.
In an analogous field of endeavor, Chan’763 teaches, wherein the raw image data of the first image volume is acquired at different bed positions of the PET system, and wherein different beta values are output for different portions of the raw image data of the first image volume acquired at the different bed positions. (Chan’763, ¶0028: “different regions within the same image can benefit from different degrees of denoising”; different regions are interpreted to represent different bed positions).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai in view of Ye, in further view of Chan and in further view of Abou Shousha using the teachings of Chan’763 to introduce region specific denoising. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of optimized denoising in different positions of an image. Therefore, it would have been obvious to combine the analogous arts Bai, Ye, Chan, Abou Shousha and Chan’763 to obtain the invention in claim 22.
Regarding claim 23, Bai in view of Chan, in further view of Abou Shousha and still in further view of Chan’763 teaches, The method of claim 22, wherein the different beta values at the different bed positions are combined to generate the recommended beta value, (Chan’763, ¶0006: “The best value for the regularization term β can depend on multiple factors, the primary of which is the application for which the reconstructed image is to be reconstructed”) and wherein the recommended beta value is used for performing the second reconstruction of the second image volume. (Chan’763, ¶0006: “modifying a reconstructed image to optimize a tradeoff between noise and resolution”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai in view of Chan, in further view of Abou Shousha and still in further view of Chan’763 using the additional teachings of Chan’763 to introduce computing the best beta value to reconstruct an image. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of optimized denoising of an image to improve image reconstruction. Therefore, it would have been obvious to combine the analogous arts Bai, Ye, Chan, Abou Shousha and Chan’763 to obtain the invention in claim 23.
Regarding claim 24, Bai in view of Chan, in further view of Abou Shousha and still in further view of Chan’763 teaches, The method of claim 22, wherein the different beta values output for different portions of the raw image data of the first image volume are individually used (Chan’763, ¶0048: “segment the image into regions, and apply regions-adaptive blending to generate a blended image in which the relative weighting between images with small and large smoothing parameters β varies by region”) to reconstruct corresponding portions of the second image in the second reconstruction. (Chan’763, ¶0033: “the small- and large-smoothing-parameter images are identical in all aspects including being reconstructed using the same IR method, except for using different values for the regularization parameter β in the cost function”).
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to modify Bai in view of Chan, in further view of Abou Shousha and still in further view of Chan’763 using the additional teachings of Chan’763 to introduce region adaptive beta values. A person skilled in the art would be motivated to combine the known elements as described above and achieve the predictable result of improving the signal to noise ratio in a reconstructed image. Therefore, it would have been obvious to combine the analogous arts Bai, Ye, Chan, Abou Shousha and Chan’763 to obtain the invention in claim 24.
Allowable Subject Matter
Claims 16, 17, 19 and 20 are allowed. The following is an examiner' s statement of reasons for indication of allowable subject matter:
Regarding independent claim 16, Bai et al. (US 2020/0334870 A1) teaches, A Positron Emission Tomography (PET) imaging system, (Bai, ¶0032: “described herein for PET imaging systems”) comprising: a processor and a non-transitory memory including instructions that when executed, cause the processor to: (Bai, ¶0004: “a non-transitory computer-readable medium stores instructions readable and executable by a workstation including at least one electronic processor to perform an image reconstruction”).
In an analogues field of endeavor, Ye teaches, a patch based denoising method that extracts noisy patches from reconstructed noisy image and updates the reconstructed image in order to reduce noise and outputs a noise regularized image.
However, none of the cited prior art references, alone or in combination, provides a motivation to teach before the effective filing date of the application, the ordered combination of: “in a first, training stage, train a customization model to learn preferences of a user of the PET imaging system based on ratings received from the user for a plurality of images, the plurality of images reconstructed using a range of noise regularization parameters, wherein the noise regularization parameters establish how close neighboring voxels in an image have to be in terms of absolute value; and in a second, inference stage: input a noise regularization parameter from the range of noise regularization parameters into the trained customization model; receive a customized noise regularization parameter as an automatic output of the trained customization model; reconstruct an image volume from raw image data using the customized noise regularization parameter; and display the image volume on a display device of the PET imaging system; and the processor and the non-transitory memory further including instructions that when executed, cause the processor to: perform a first reconstruction of a first image volume using raw image data of the PET system without specifying any noise regularization parameters, extract a plurality of patches from the first reconstruction of the first image volume, use a prediction model of the PET imaging system to automatically predict a noise regularization parameter setting based on the patches extracted from the first reconstruction of the first image volume, input the predicted noise regularization parameter setting output by the prediction model into the trained customization model to automatically customize the predicted noise regularization parameter setting, and perform a second reconstruction of a second image volume using the raw image data of the PET system and the automatically customized, predicted noise regularization parameter setting, the second image volume having a higher image quality than the first image volume.” with the other limitations recited in the independent claim 1. Dependent claims 17, 19 and 20 are also allowable based on their respective dependencies from allowable independent base claim 16.
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 MEHRAZUL ISLAM whose telephone number is (571)270-0489. The examiner can normally be reached Monday-Friday: 8am-5pm.
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, Saini Amandeep can be reached at (571) 272-3382. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of 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.
/MEHRAZUL ISLAM/Examiner, Art Unit 2662
/AMANDEEP SAINI/Supervisory Patent Examiner, Art Unit 2662
Prosecution Insights