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 . This action is responsive to Applicant's amendment and remarks filed June 24, 2026. Claims 1-20 are pending . Claims 3, 5, 8, 13, 15, and 18 were amended; the independent claims were not amended. THIS ACTION IS MADE FINAL.
Response to Arguments
Applicant's arguments filed June 24, 2026 have been fully considered but they are not persuasive.
Applicant argues that “Siemens does not disclose or suggest a dynamic filter,” that Zhang “does not disclose or suggest a dynamic filter,” and that “Groff similarly” does not disclose the recited limitations.
Nonobviousness cannot be established by attacking references individually where the rejection is based on a combination of references. In re Keller, 642 F.2d 413 (CCPA 1981): In re Merck & Co., 800 F.2d 1091 (Fed. Cir. 1986). The “dynamic filter” of claims 1 and 11 is not mapped to SIEMONS; it is mapped to ZHANG.
Applicant's contention that ZHANG's stabilization “does not disclose or suggest a dynamic filter” does not address the claim construction on which the rejection rests. Under the broadest reasonable interpretation in light of the specification, a “dynamic filter” reads on an operation that stabilizes the sequential wavefront measurement stream by combining measurements over time, as set forth above. The specification describes the dynamic filter as a Kalman filter that “recursively combin[es] wavefront measurements” (¶ 41). The claimed “dynamic filter” therefore reads on an operation that stabilizes the sequential wavefront measurement stream by combining measurements over time
ZHANG's averaging of 100-300 subregions to stabilize the measurements for “tracking dynamic wavefront distortions during continuous data acquisition” (Results) reads on the operation dynamic filter’s operation from the specification. Applicant has not proposed a construction of “dynamic filter” that would shows this is an error. Temporal combination of a sequential measurement stream to stabilize its output before actuation is a filtering operation on a dynamic signal, that is, a dynamic filter.
Applicant does not separately argue the dependent claims or the rejection of claims 10 and 20 over SIEMONS, ZHANG, and GROFF. The rejections are maintained.
Therefore, the argued limitations were written broad such that they read upon the cited references or are shown explicitly by the references. As a result, the claims stand as follows.
Claim Rejections -35 U.S.C. § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 3 and 13 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Amended claims 3 and 13 are rejected under 35 U.S.C. 112(b) as failing to particularly point out
and distinctly claim the subject matter regarded as the invention.
Amended claims 3 and 13 recite that the processor “restores single molecule emission patterns to pre-analysis conditions.” Deletion of the word “approaching” does not cure the indefiniteness. The specification quantifies PSF restoration by normalized cross-correlation (NCC) values of 0.95±0.02 and residual Wrms of 0.29±0.12 rad (¶ 43), but does not define what the “pre-analysis conditions” to which the emission patterns are said to be restored, or the threshold when the restoration to that state is reached as distinguished from failing to reach it. The spec reports only partial restoration, it does not show restoration to the pre-aberration state. A person of ordinary skill in the art would not be informed of the metes and bounds of this limitation with reasonable certainty. See Nautilus, Inc. v. Biosig Instruments, Inc., 572 U.S. 898 (2014). Applicant was invited to amend to a specific quantitative threshold, e.g., an NCC value of at least 0.95, consistent with the specification, and declined. This rejection is maintained.
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 1-9 and 11-19 are rejected under 35 U.S.C. 103 as being unpatentable over Siemons et al., "Robust adaptive optics for localization microscopy deep in complex tissue," Nature Communications, Vol. 12, Art. 3407 (2021) (hereinafter "SIEMONS") in view of Zhang et al., "Analyzing complex single-molecule emission patterns with deep learning," Nature Methods, Vol. 15, pp. 913- 916 (2018) (hereinafter "ZHANG").
The Examiner notes that reference "ZHANG" identified below is a self-disclosure by named inventors Fang Huang and Peiyi Zhang, who are co-authors of that publication. ZHANG
was published in October 2018, which is more than one year prior to the effective filing date of
May 30, 2023. The 35 U.S.C. 102(b)(1)(A) grace period exception does not apply, and ZHANG
therefore constitutes prior art under 35 U.S.C. 102(a)(1) that cannot be antedated by affidavit.
Claims 1, 11, and 20.
SIEMONS and ZHANG disclose a system comprising: an imaging apparatus configured for
conducting single-molecule localization microscopy (SMLM), wherein the imaging apparatus
comprises a deformable mirror (SIEMONS: "Model-based optimization iteratively corrects Zernike modes by applying a sequence of biases of the Zernike mode that is to be corrected. The metric values of these acquisitions are computed and a metric curve is fitted to find the optimum" (Methods: AO methods); and "MICAO adaptive optics module containing a MIRAO52E (Imagine Optics) deformable mirror was mounted" (Methods: Set up). This teaches an imaging apparatus for SMLM comprising a deformable mirror controlled via automated metric-based optimization.) and (ZHANG: " these wavefront measurements made with smNet stabilized after averaging of 100–300 subregions, or 20–60 raw data frames" (Results p. 915). This teaches a dynamic filter that is temporal aggregation of sequentially noisy measurement to produce a smoothed/averaged output. This sits structurally between the per-frame smNet output and the actuator that moves the deformable mirror.); and
a processor operably associated with the imaging apparatus and configured to:
monitor individual emission patterns produced via the imaging apparatus from a plurality of
different single molecules in a sample (SIEMONS: "REALM ... corrects aberrations of up to 1 rad RMS using 297 frames of blinking molecules to improve single-molecule localization" (Abstract). This teaches a processor monitoring individual emission patterns from a plurality of blinking single molecules to evaluate and correct sample-induced aberrations.);
(not specifically taught by SIEMONS, which employs iterative image-quality metrics rather than direct neural network inference of wavefront distortion);
provide the shared wavefront distortion for each of the individual emission patterns through the dynamic filter (ZHANG: "These wavefront measurements made with smNet stabilized after averaging of 100-300 subregions, or 20-60 raw data frames" (Results p. 915). This teaches providing sequential wavefront distortion estimates through a dynamic filter that combines measurements over time to stabilize the output before mirror correction.); and
operate the deformable mirror to compensate for sample induced aberrations (SIEMONS: "REALM enables to resolve the periodic organization of cytoskeletal spectrin of the axon initial segment even at 50 µm depth in brain tissue " (Abstract). This teaches operating a deformable mirror to compensate for sample-induced aberrations. The present specification at ¶ 44 further admits that SIEMONS' method "requir[es] a total of 330 updates to compensate 11 aberration types," confirming SIEMONS teaches this limitation.).
SIEMONS does not specifically teach "a dynamic filter" and "inferring shared wavefront distortion" via a deep neural network applied to individual emission patterns. However, ZHANG, in the same field of endeavor, teaches a dynamic filter (shown above) and inferring shared wavefront distortion from individual single molecule emission patterns (ZHANG: "Although the emission patterns in a single-molecule dataset originate from different locations within the detected region of the specimen, they share a similar wavefront distortion induced by the inhomogeneous refractive indices of cell and tissue structures. smNet ... directly extracted the shared wavefront distortion from a small collection of detected emission patterns without any additional information" (Results p. 914). This teaches a processor configured to infer shared wavefront distortion for each of the individual emission patterns.).
Therefore, it would have been obvious to one of ordinary skill in the art to combine
SIEMONS and ZHANG before the effective filing date of the claimed invention. The motivation for this combination would have been to replace SIEMONS's slow iterative metric-based aberration sensing, which requires up to 330 mirror updates to compensate 11 aberration modes (as admitted at [0044]), with ZHANG's direct neural network inference of shared wavefront distortion, enabling real-time, fewer-update correction during live SMLM experiments. ZHANG demonstrated that smNet tracks dynamic wavefront changes within 20-60 raw data frames (Fig. 3e), directly establishing feasibility of integrating ZHANG's inference capability into SIEMONS's mirror-control loop. Both references address the same technical problem of correcting sample-induced aberrations in SMLM using the blinking emission patterns of single molecules as the sensing signal.
Claim 2 and 12.
Claims 2 and 12 further recite that the processor simultaneously estimates and compensates for 28 types of wavefront deformation shapes. ZHANG teaches a neural network configured to
simultaneously estimate amplitudes of multiple wavefront aberration modes (ZHANG: "In our
study, the output of smNet was a vector of 12 or 21 elements representing the amplitudes of
12 or 21 Zernike modes" (Methods). This teaches simultaneously estimating and correcting
multiple types of wavefront deformation shapes via a DNN output vector.) The specific
number of simultaneously corrected modes-here, 28-is governed by the actuator count and
chosen correction basis of the deformable mirror hardware in use, and represents a routine design optimization within the skill of a person of ordinary skill in the art configuring a DNN-based adaptive optics system for a particular mirror platform.
Claims 3 and 13.
Claims 3 and 13 further recite that the processor restores single molecule emission patterns
approaching pre-analysis conditions. SIEMONS and ZHANG together teach AO-corrected
restoration of aberrated PSFs toward their diffraction-limited pre-aberration state. SIEMONS
teaches that REALM restores the PSF to near-diffraction-limited quality, enabling localization
precisions below 20 nm (SIEMONS: "With REALM, the PSF is more symmetric and remains focused along the focal depth of the objective. This increases the detection of dim blinking events as well as the detection of molecules inside the complete depth of focus" (Results p. 5). This teaches restoring single-molecule emission patterns toward their preaberration, diffraction-limited form.) ZHANG teaches that smNet achieves wavefront correction with high fidelity, such that the corrected PSF closely matches the pre-aberration ground truth (ZHANG: "smNet was capable of simultaneously measuring amplitudes of 12-21 Zernike polynomials (Wyant order), representing wavefront shapes, while achieving a residual wavefront error of <30 mλ" (Results p. 914). This teaches that the DNN inference enables restoration of emission patterns approaching pre-analysis conditions by reducing residual wavefront error to near-zero.)
Claims 4 and 14.
Claims 4 and 14 further recite improving resolution and fidelity of 3D SMLM through tissue
specimens over 130 micrometers. SIEMONS teaches AO-corrected SMLM imaging at depths up
to 50 μm in brain tissue, demonstrating feasibility up to 80 μm with spherical aberration precorrection, and explicitly anticipates deeper imaging (SIEMONS: "REALM ... enabling robust
SMLM at 50 μm depth and even up to 80 μm depth when pre-correcting spherical aberration" (Introduction); "We anticipate that REALM can enable 80 or 100 μm deep imaging in tissue when combined with" reduced-background approaches (Discussion). This teaches progressive depth extension achievable with system optimization.)
Claims 5 and 15.
Claims 5 and 15 further recite that the improvement is accomplished in as few as 3-20 mirror
changes. ZHANG demonstrates that smNet's wavefront estimates stabilize over a small number
of acquisitions (ZHANG: "These wavefront measurements made with smNet stabilized after
averaging of 100-300 subregions, or 20-60 raw data frames" (Results). This teaches that
DNN-based wavefront inference converges rapidly, enabling mirror correction based on a
minimal number of acquisitions.) In the proposed combination, ZHANG's direct DNN inference replaces SIEMONS's iterative probe-and-measure approach, which requires 297 acquisitions across three correction rounds. Reducing the number of required mirror commands to a range of 3-20 is a result-effective variable. This is a design optimization of iteration count and convergence threshold, within the ordinary skill in the art.
Claims 6 and 16.
Claims 6 and 16 further recite the processor is trained via a training data set comprising
segmenting single molecule-containing sub-regions. ZHANG explicitly teaches processing
individual PSF subregions as the core input unit of the smNet workflow (ZHANG: "The complex and subtle features of the PSF lie in the photon distribution within a small subregion (2-10 μm2)" (Results); and "These wavefront measurements made with smNet stabilized after averaging of 100-300 subregions" (Results). This teaches training and operating the processor using a training dataset comprising segmented single molecule containing sub-regions.)
Claims 7 and 17.
Claims 7 and 17 further recite each sub-region goes through a sequence of template matching
processes organized as convolutional layers and residual blocks with PReLU activations and
batch normalizations in between. ZHANG explicitly teaches this architecture (ZHANG: "smNet
is a deep network of 27-36 layers (Supplementary Figs. 1-3 and Supplementary Table 1)
consisting of convolutional layers, residual blocks, and fully connected layers together with
batch normalization and parametric rectified linear unit" (Results p. 913); and "Each convolutional layer is followed by batch normalization and a parametric rectified linear
unit (PReLU)" (Methods: smNet architecture). This teaches the claimed sequence of convolutional layers and residual blocks with PReLU activations and batch normalizations.)
Claims 8 and 18.
Claims 8 and 18 further recite the processor fully connects through 1 x 1 convolutional layers to
an output vector of amplitude estimates for wavefront shapes in terms of the native mirror deformation modes. ZHANG teaches fully connecting through 1 x 1 convolutional layers to a
vector of amplitude estimates for wavefront shapes (ZHANG: "We used 1 x 1 convolutional
layers. Finally, we followed these by fully connected layers ... the output of smNet was a
vector of 12 or 21 elements representing the amplitudes of 12 or 21 Zernike modes"
(Methods: smNet architecture). This teaches the 1x1 convolutional output layer producing
a vector of amplitude estimates for wavefront shapes.)
Claims 9 and 19.
Claims 9 and 19 further recite the wavefront is represented with coefficients of orthogonal basis. ZHANG teaches representing the wavefront using Zernike polynomials, which constitute a well-established orthogonal basis for optical wavefront description (ZHANG: "smNet was capable of simultaneously measuring amplitudes of 12-21 Zernike polynomials (Wyant order),
representing wavefront shapes" (Results); and "the output of smNet was a vector of 12 or
21 elements representing the amplitudes of 12 or 21 Zernike modes" (Methods). Zernike
polynomials are a well-known orthogonal basis for representing optical wavefronts. This
teaches representing the wavefront with coefficients of an orthogonal basis.)
Claims 10 and 20 are rejected over SIEMONS and ZHANG and further in view of "Kalman Filter Estimation for Focal Plane Wavefront Correction" (hereinafter "GROFF").
Claims 10 and 20.
Claims 10 and 20 further recite the dynamic filter is a Kalman filter. GROFF teaches applying a Kalman filter as an optimal recursive estimator for focal-plane wavefront estimation and control, combining sequential noisy wavefront measurements to produce a stabilized state estimate before commanding a physical actuator (GROFF: “We demonstrate a Kalman filter estimator that uses prior knowledge to create the estimate of the electric field, dramatically reducing the number of exposures required to estimate the image plane electric field" (Abstract); and “In doing so we do want to consider the relative effect of process and detector noise to optimally combine an extrapolation of the state estimate with new measurement updates. This is exactly the problem a discrete time Kalman filter solves” (§5 p. 5) This teaches applying a Kalman filter as a dynamic filter that recursively combines prior wavefront state estimates with new noisy measurements to produce stabilized output for deformable mirror actuation.) In the proposed combination, ZHANG demonstrates that raw smNet wavefront estimates require averaging over 100-300 subregions to stabilize (ZHANG: "These wavefront measurements made with smNet stabilized after averaging of 100-300 subregions, or 20-60 raw data frames" (Results).) A person of ordinary skill in the art would have recognized that applying a Kalman filter, the canonical linear-Gaussian recursive estimator, to ZHANG's sequential DNN wavefront outputs before commanding SIEMONS's deformable mirror actuator would yield predictably improved estimation stability. The substitution of a Kalman filter for simple temporal averaging yields no change in the underlying DNN or AO architecture and would have been an obvious design choice.
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 Ross Varndell whose telephone number is (571)270-1922. The examiner can normally be reached M-F, 9-5 EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, O’Neal Mistry can be reached at (313)446-4912. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see https://ppair-my.uspto.gov/pair/PrivatePair. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/Ross Varndell/Primary Examiner, Art Unit 2674
Prosecution Insights