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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 7/18/2025 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
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.
Claim(s) 1-5 are rejected under 35 U.S.C. 103 as being unpatentable over Dong et al. (Hereinafter, “Dong”) in the US Patent Application Publication Number US 20220326329 A1 in view of Sun et al. (Hereinafter, “Sun”) in the US Patent Application Number US 20200363485 A1.
Regarding claim 1, Dong teaches method for magnetic resonance imaging (a method for performing magnetic resonance image reconstruction based on mixed-trajectory acquired and recorded magnetic resonance signal data, combining parallel imaging and compressed sensing reconstruction; Paragraph [0002] Line 2-6), the method comprising:
a) performing by an MRI scanner (Figure 2) an MRI scan to acquire non-Cartesian k-space [5e] in Figure 1 MRI acquisition data (the MRI imaging method is adapted to perform MRI imaging of k-space acquired in mixed trajectories, wherein the k-space acquired in mixed trajectories can have the acquired MR signals occupy a central region of k-space with full acquisition along a Cartesian trajectory, and occupy a peripheral region of the k-space with under-acquisition along a non-Cartesian trajectory; Paragraph [0068] Line 27-34; Referring to FIG. 3, in step S1, MR signal data acquired by the at least one receiving coil 3 occupies the central region 5d of the first k-space 5 along the Cartesian trajectory 5a under gradient magnetic field control with full sampling, and acquired MR signal data occupies the peripheral region 5e of the first k-space 5 along the non-Cartesian trajectory under gradient magnetic field control with undersampling; Paragraph [0070] Line 1-8);
b) estimating by the MRI scanner Cartesian k-space [5a] data (It must be explained that FIG. 1 shows a 2D projection of acquisition trajectories for obtaining the first k-space 5 based on execution of the PETRA pulse sequence, wherein the circle of dots represents k-space acquired along the Cartesian trajectory 5a; Paragraph [0065] Line 1-5) from the non-Cartesian k-space MRI acquisition data (In the reconstruction of first image data, first of all, for example by means of an inverse Fourier transform, the extracted data occupying the non-Cartesian trajectories 5b in the peripheral region 5e of the first k-space 5 is transformed to the image domain to obtain first image data, then a filter transform is applied to the first image data and the transformation thereof into a sparse representation transform domain is performed, and artifacts introduced due to undersampling in the first image data (the non-Cartesian trajectory part) are suppressed during image reconstruction. For this purpose, when performing image reconstruction of the first image data, it is necessary to obtain a solution satisfying a constraint condition, for example such that the first image data under the action of the filter transform is minimized under the norm l.sub.1, and such that a regularization constraint is satisfied between measured k-space and k-space under the Fourier transform in the process of reconstructing the first image data, e.g. is a minimum under the norm l.sub.2. Here, the reconstructed first image data is reconstructed with the aid of compressed sensing, i.e. with the aid of compressed sensing, the sparse representation of the first image data in the transform domain is used to perform suppression of artifacts caused by undersampling in a first image during image reconstruction, and the second k-space can be obtained by applying a Fourier transform to the reconstructed first image data; here, compressed sensing reconstruction can be suitable for iteration; Paragraph [0073] Line 1-27),
wherein the estimating comprises estimating each Cartesian k-space coordinate in the Cartesian k-space data from multiple neighboring non-Cartesian k-space coordinates (In step S2, based on extraction of data occupying the non-Cartesian trajectories 5b in the peripheral region 5e of the first k-space 5, a second k-space is constructed by reconstructing first image data, wherein reconstructing the first image data at least comprises using a sparse representation of the first image data in a transform domain to perform suppression of artifacts caused by the undersampling in the first image data during image reconstruction, and wherein data occupies a Cartesian trajectory in the second k-space obtained by transformation based on the reconstructed first image data. Here, the reconstructed first image data may obtain the second k-space under a Fourier transform or fast Fourier transform, and data thereof is recorded in the second k-space by occupying a Cartesian trajectory; Paragraph [0072] Line 1-14; In step S3, based on synthesis of the second k-space with data occupying the Cartesian trajectory 5a in the central region 5d of the first k-space 5, a k-space suitable for MRI is generated. Specifically, based on extraction of data occupying the Cartesian trajectory 5a in the central region 5d of the first k-space 5 to replace data occupying the Cartesian trajectory in the corresponding central region in the second k-space, the k-space suitable for MRI can be generated; Paragraph [0074] Line 1-9) using an ensemble of GRAPPA kernels (Acquisition is accelerated by undersampling k-space and using sensitivity information for image reconstruction. Here, there are two main technical paths: SENSE (SENSitivity Encoding), based on explicit coil sensitivity; and GRAPPA (GeneRalized Autocalibrating Partial Parallel Acquisition), based on the use of known relevance in k-space. ESPIRiT (Iterative self-consistent parallel imaging reconstruction using eigenvector maps) is an eigenvector-based autocalibration technique, which combines the advantages of SENSE and robustness for specific errors similar to GRAPPA; Paragraph [0060] Line 7-16; Here, the sensitivity distribution information of the receiving coils 3 may be calculated with the aid of a method such as SENSE, GRAPPA, SMASH, AUTO-SMASH or SPIRiT; Paragraph [0080] Line 1-4),
c) reconstructing by the MRI scanner an MRI image from the estimated Cartesian k-space data (In step S40, based on the sensitivity distribution information, first image data is reconstructed for data occupying the non-Cartesian trajectories in the peripheral region 5e of the first k-space 5 in order to construct a second k-space, wherein reconstructing the first image data comprises performing parallel imaging with the aid of the sensitivity distribution information and using sparsity or a sparse representation of the first image data in a transform domain to perform suppression of artifacts caused by undersampling in the first image data during image reconstruction in order to obtain reconstructed first image data, and wherein data occupies a Cartesian trajectory in the second k-space and is recorded; Paragraph [0084] Line 1-13).
However, Dong fails to teach where each of the GRAPPA kernels is obtained from a
non-linear model trained on calibration data from an MRI calibration scan.
Sun teaches magnetic resonance imaging (MRI) operations that, according to some aspects, reconstructs MRI images from spirally acquired MRI data and provides final images having suppressed artifact images across numerous imaging domains (Paragraph [0003] Line 1-5),
where each of the GRAPPA kernels is obtained from a non-linear model trained on calibration data from an MRI calibration scan (FIG. 6 is a flow chart of one method of reconstructing image data from acquired MRI data. FIG. 6 shows reconstruction of MB spiral data using spiral slice-GRAPPA. (A) The non-Cartesian slice-GRAPPA method applied to MB spiral k-space data uses phase demodulation of the sth slice, followed by gridding and application of the slice-GRAPPA kernel of the sth slice, as shown in Eq. 5 below. In a final step, the IFFT is performed; Paragraph [0016] Line 1-8; FIG. 14 FIG. 14A is a flow chart of one method of reconstructing image data from acquired MRI data. Separation of multiband (MB) spiral k-space by spiral slice-GRAPPA. For kernel calibration, single-band data of the center of k-space are acquired for all slices, phase modulation is applied to all slices, and phase demodulation corresponding to the sth slice is applied to all slices; Paragraph [0027] Line 1-7). The purpose of doing so is to form a calibrated slice of k-space multi-band image data; converting the calibrated slice of multi-band k-space image data to an output image, to calibrate an SG kernel and separate MB data, there is an opportunity to additionally use in-plane coil sensitivity information to calibrate an in-plane SPIRiT kernel and to develop an iterative reconstruction model that enforces in-plane coil consistency, through-plane coil consistency, and consistency with the acquired MB data.
It would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention, to modify Panetta and Dong in view of Sun to obtain GRAPPA kernels, because Sun teaches to obtain GRAPPA kernels from a non-linear model trained on calibration data from an MRI calibration scan, forms a calibrated slice of k-space multi-band image data; converting the calibrated slice of multi-band k-space image data to an output image (Paragraph [0008]), calibrates an SG kernel and separate MB data, there is an opportunity to additionally use in-plane coil sensitivity information to calibrate an in-plane SPIRiT kernel and to develop an iterative reconstruction model that enforces in-plane coil consistency, through-plane coil consistency, and consistency with the acquired MB data (Paragraph [0104]).
Regarding claim 2, Dong teaches a method,
wherein the non-linear model (In addition, when the extracted data occupying the non-Cartesian trajectories, e.g. the radial trajectories 5b, in the peripheral region 5e of the first k-space 5 is transformed to the sparse representation transform domain, the filter transform or non-linear filter transform (e.g. linear operator or non-linear operator) applied may for example use a sparse transform such as a small-wave transform, a discrete small-wave transform, a discrete cosine transform or a finite difference transform; Paragraph [0086] Line -12) is a neural network (In addition, it is also possible to use for example deep learning or a neural network to reconstruct the first image data; Paragraph [0088] Line 1-3).
Regarding claim 3, Dong teaches a method,
wherein the non-linear model (In addition, when the extracted data occupying the non-Cartesian trajectories, e.g. the radial trajectories 5b, in the peripheral region 5e of the first k-space 5 is transformed to the sparse representation transform domain, the filter transform or non-linear filter transform (e.g. linear operator or non-linear operator) applied may for example use a sparse transform such as a small-wave transform, a discrete small-wave transform, a discrete cosine transform or a finite difference transform; Paragraph [0086] Line 1-12) is a kernel regression model (In step S201, gridding processing is applied to a first k-space 5. Here, for example, grid kernel function convolution is applied to the first k-space to achieve gridding of the first k-space 5; Paragraph [0094] Line 1-4).
Regarding claim 4, Dong teaches a method,
wherein the non-linear model is a linear model with heuristically chosen non-linear features lifting (In addition, when the extracted data occupying the non-Cartesian trajectories, e.g. the radial trajectories 5b, in the peripheral region 5e of the first k-space 5 is transformed to the sparse representation transform domain, the filter transform or non-linear filter transform (e.g. linear operator or non-linear operator) applied may for example use a sparse transform such as a small-wave transform, a discrete small-wave transform, a discrete cosine transform or a finite difference transform; Paragraph [0086] Line 1-12).
Regarding claim 5, Dong teaches a method,
wherein the non-linear features lifting comprises sine and cosine positional encoding (In addition, when the extracted data occupying the non-Cartesian trajectories, e.g. the radial trajectories 5b, in the peripheral region 5e of the first k-space 5 is transformed to the sparse representation transform domain, the filter transform or non-linear filter transform (e.g. linear operator or non-linear operator) applied may for example use a sparse transform such as a small-wave transform, a discrete small-wave transform, a discrete cosine transform or a finite difference transform; Paragraph [0086] Line 1-12; therefore finite discrete transform comprises sine and cosine positional encoding).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
GRISWOLD et al. (US 20100201363 A1) discloses, “CALIBRATING PARALLEL MRI WITH CARTESIAN CONTINUOUS SAMPLING-[0017] systems and methods may calibrate a GRAPPA reconstruction using both Cartesian continuous sampling and ACS lines. [0021] FIG. 1 illustrates a portion 100 of a pulse sequence having an extended acquisition window 110, and an overlapping phase-encoding gradient 130 and read gradient 140. The overlapping phase-encoding gradients 130 and read gradients 140 yield a trajectory having both Cartesian and non-Cartesian segments. FIG. 2 illustrates a trajectory 200 having both Cartesian segments (e.g., segment 210, segment 220) and non-Cartesian segments (e.g., segment 230, segment 240). Returning now to FIG. 1, note the extended acquisition window 110 as compared to a conventional acquisition window 120. Extending the acquisition window facilitates acquiring additional data that can be used for calibration. The non-Cartesian portion of the resulting trajectory coupled with additional sampling causes central k-space to be over-sampled with respect to the desired under sampling. The non-Cartesian portions (e.g., 230, 240) represent the time during which a traversal is being made from one k-space line (e.g., 220) to another k-space line (e.g., 210). While the extended acquisition window 110 is illustrated encompassing all of the portion 100 of the pulse sequence, it is to be appreciated that the extended acquisition window may cover less than all of portion 100. [0022] In one example, the non-Cartesian portion of the resulting trajectory leads to central k-space being sampled in a manner that satisfies the Nyquist criterion. Thus, this critically and over-sampled data may be used to calibrate a reconstruction process. In one example the reconstruction process may be a GRAPPA reconstruction process. After the calibration is performed, data in an under-sampled Cartesian area can be reconstructed based, at least in part, on the data acquired in the region that satisfies the Nyquist criterion and the calibration data-However GRISWOLD does not disclose estimating by the MRI scanner Cartesian k-space data from the non-Cartesian k-space MRI acquisition data, wherein the estimating comprises estimating each Cartesian k-space coordinate in the Cartesian k-space data from multiple neighboring non-Cartesian k-space coordinates using an ensemble of GRAPPA kernels, where each of the GRAPPA kernels is obtained from a non-linear model trained on calibration data from an MRI calibration scan.”
Any inquiry concerning this communication or earlier communications from the examiner should be directed to NASIMA MONSUR whose telephone number is (571)272-8497. The examiner can normally be reached 10:00 am-6:00 pm.
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, Eman Alkafawi can be reached at (571) 272-4448. 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.
/NASIMA MONSUR/Primary Examiner, Art Unit 2858
Prosecution Insights