Prosecution Insights
Last updated: November 20, 2026
Application No. 18/735,144

COMBINING DEEP LEARNING MODEL HIDDEN LAYER OUTPUT WITH SPECIMEN-SPECIFIC INPUT FOR DEFECT CLASSIFICATION OR ANOTHER SEMICONDUCTOR APPLICATION

Non-Final OA §102§103§112
Filed
Jun 05, 2024
Priority
Apr 25, 2024 — provisional 63/638,413
Examiner
BOYAR, NOAH WILLIAM
Art Unit
2669
Tech Center
2600 — Communications
Assignee
KLA Corporation
OA Round
3 (Non-Final)
100%
Grant Probability
Favorable
3-4
OA Rounds
0m
Est. Remaining
99%
With Interview

Examiner Intelligence

Grants 100% — above average
100%
Career Allowance Rate
10 granted / 10 resolved
+38.0% vs TC avg
Minimal +0% lift
Without
With
+0.0%
Interview Lift
resolved cases with interview
Typical timeline
2y 6m
Avg Prosecution
25 currently pending
Career history
39
Total Applications
across all art units

Statute-Specific Performance

§101
16.4%
-23.6% vs TC avg
§103
41.4%
+1.4% vs TC avg
§102
17.2%
-22.8% vs TC avg
§112
22.7%
-17.3% vs TC avg
Black line = Tech Center average estimate • Based on career data from 10 resolved cases

Office Action

§102 §103 §112
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 . Continued Examination Under 37 CFR 1.114 A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 9/27/2026 has been entered. Response to Amendment The amendment filed on 9/27/2026 has been entered. Claims 1-20 remain pending in the application. Response to Arguments/Remarks re. 35 U.S.C. § 103 Rejections Arguments have been fully considered but are moot as the new grounds of rejection does not rely on the previously cited references for any teaching specifically challenged. Claim Rejections - 35 USC § 112(a) The following is a quotation of the first paragraph of 35 U.S.C. 112(a): (a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention. Claims 1-20 are rejected under 35 U.S.C. 112(a) as failing to comply with the written description requirement. The claims contain subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor at the time the application was filed, had possession of the claimed invention. With respect to claim 1, the claim recites: and wherein the hidden layer output is not the information determined for the specimen “Any negative limitation or exclusionary proviso must have basis in the original disclosure”. See MPEP § 2173.05(i). As amended subject matter, the examiner looks to the specification for any explicit establishment that “the hidden layer output is not the information determined for the specimen” (Page 22): PNG media_image1.png 516 748 media_image1.png Greyscale As a non-limiting example of an embodiment of the claimed invention, there is insufficient support to provide that the “hidden layer output” cannot be the “information determined for the specimen”. The specification merely states: “For example, hidden layer output is generally not useful to users in the same way specimen information generated by the output layer is” “the hidden layer output is not in of itself a defect classification” “…the hidden layer output is not important to users and is not, in of itself, information that is useful for users” The specification did not state that this hidden layer output itself is not “information determined for the specimen” (in a computer system) – only that it is not useful for users. It is contradicting that a hidden layer output can be used to: “determine a defect classification, which is output by the output layer…” “…while the hidden layer output is useful for determining specimen information such as defect classifications…” While being of so little value that it cannot even be referred to under the broad label of “information determined for the specimen”. Pages 29-30 further weighs against explicit establishment that the hidden layer output cannot be information determined for the specimen: “In some embodiments, the hidden layer output generated by the at least one of the hidden layers includes multiple different results, the input specific to the specimen includes multiple different inputs, and the additional component is configured for independently applying an importance to at least one of the multiple different results and the multiple different inputs prior to determining the additional information. The importance of the different results and the different inputs may be defined relative to other results and/or inputs, and so the importance may be a kind of relative importance of one input or result to another. The importances may be determined as part of the learning process of the RF that assigns such importances. The importances assigned to each of the results and/or inputs may also be a kind of weight that is learned as described above and then applied to the results and/or inputs during classification” This paragraph characterizes hidden layer output as not simply a completely useless intermediate input, but itself a “result” that can be weighed by importance relative to other “results”. Similarly, page 24 also implies that this hidden layer output can be at least thought of as information: “The component(s) also include an additional component configured for determining additional information for the specimen from the hidden layer output generated by at least one of the hidden layers in combination with input specific to the specimen…where the output of a hidden layer of the DL model is used as input to the RF along with other information described further herein such as defect attributes, which may be determined or derived from the specimen regions, image segmentation, detection algorithm settings, etc” As such, page 22’s recitation that: “…the inventors have created a new way to use the hidden layer output to improve specimen information determination processes” Does not exclude a reading that certain respective hidden layer outputs cannot be “the information” and/or the building blocks of the information. Such that the sentence could permissibly be read as: “using hidden layer outputs (in combination) to generate improved specimen information (by at least one final hidden layer)” Such a reading is also consistent with disclosure of page 26: “Although the output of the final fully connected layer in the DL model may be most suitable for use as the hidden layer output in the embodiments described herein, the output of this layer and only this layer does not need to be used in the embodiments described herein… the hidden layer output used in the embodiments described herein may include output generated by more than one intermediate layer in the same DL model. For example, one suitable DL model configuration may include a set of convolution layers that “concentrate” the images down to a smaller number of features that are then fed into fully connected layers. It is conceivable that the hidden layer output of one or more of such convolution layers may be input with the hidden layer output of one or more of the fully connected layers into the additional component as described further herein” To the extent Applicant has argued that “features extracted” do not constitute “information”, the specification is understood to be non-limiting. The specification specifically considered embodiments where the DL model is configured to other tasks (page 26): “For example, a DL model that makes some decision regarding an input, e.g., classifying an image as containing a defect or not, classification of an image as an image of one type of defect or another, may in general include at least one fully connected layer. However, the DL model may be configured for a different function, such as transforming an image from one resolution to another or transforming one image type to a different image type” As such, the DL model as disclosed does not affirmatively exclude a task which necessarily involves hidden layer(s) extracting “information” (however the Applicant may define the term). Accordingly, claim 1 is rejected as reciting new matter. Claims 19 and 20 are rejected as they recite in parallel the new matter of claim 1. Claims 2-18 are rejected by virtue of their dependency on the new matter of claim 1. Claim Rejections - 35 USC § 112(b) 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. Claim 17 is rejected under 35 U.S.C. 112(b) as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention. With respect to claim 17, the claim recites: “wherein the hidden layer output generated by the at least one of the hidden layers comprises multiple different results” It is not clear whether one layer must generate multiple different results (under the “one” hidden layer embodiment of the claim), or if under the recitation of “at least one”, the claim language can also be satisfied by multiple different results being generated (in sum) by the contributions of multiple different hidden layers. For example: wherein the hidden layer output generated by the at least one of the hidden layers (i.e., three hidden layers) comprises multiple different results (result A for layer 1, result B for layer 2, result C for layer 3). Additionally, the claim also recites: and wherein the additional component is further configured for independently applying an importance to at least one of the multiple different results and the multiple different inputs prior to determining the additional information It is also unclear whether the “at least one” refers to: and wherein the additional component is further configured for independently applying an importance to at least one of (the multiple different results) and (the multiple different inputs) prior to determining the additional information Or: and wherein the additional component is further configured for independently applying an importance to at least one of the multiple different results (at least one of the results) and the multiple different inputs prior to determining the additional information Or: and wherein the additional component is further configured for independently applying an importance to at least one of the multiple different results (at least one of the results) and the multiple different inputs prior (at least one of the different inputs) to determining the additional information Accordingly, the language is indefinite. In the interest of compact prosecution, all interpretations outlined will be considered for the application of prior art. Claim Rejections - 35 USC § 102 The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action: A person shall be entitled to a patent unless – (a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention. (a)(2) the claimed invention was described in a patent issued under section 151, or in an application for patent published or deemed published under section 122(b), in which the patent or application, as the case may be, names another inventor and was effectively filed before the effective filing date of the claimed invention. Claims 1-2, 6-8, 10-13, and 17-20 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Dong et. al Defect Detection and Classification by Training a Generic Convolutional Neural Network Encoder (Hereinafter, “Dong”). With respect to claim 1, Dong discloses: A system configured for determining information for a specimen, comprising a computer system configured for acquiring output generated for a specimen ([3] “To derive more training images with the pixel-level annotation data, we use an image fusion technique to transfer each defect contained in a publicly available abnormal dataset to a real normal image that we obtained from an aerospace manufacturing factory”; ([4A] “In all experiments, an NVIDIA Titan XP graphics card was used together with a Lenovo P520 workstation”) by one or more detectors of an output generation subsystem ([6A(1)] “This dataset was acquired by Dong et al. [4] in the factory of an aerospace manufacturing company, where a digital radiography device designed by Bosello was used to acquire the images”) one or more components executed by the computer system, wherein the one or more components comprise a deep learning model configured for determining information for the specimen from the output generated by at least one of the one or more detectors (Fig. 4, “U-NET Training and Prediction”), wherein the output from which the information is determined comprises images of the specimen (Fig. 4; [4]), wherein the deep learning model comprises hidden layers configured for generating hidden layer output (Fig, 4, “Encoder”), and wherein the hidden layer output is not the information determined for the specimen (Fig. 4, “Softmax”; negative limitation satisfied as softmax layers are used to “determine[] information” (prediction); [4C(1)] “After fine-tuning was complete, the softmax of the U-Net model was used to produce a pixel-level prediction”) an additional component configured for determining additional information for the specimen from the hidden layer output generated by at least one of the hidden layers (Fig. 4, “Detection by Random Forests”) in combination with input specific to the specimen (Fig. 4, “Interpolation”, resulting in new input-specific features that are also sent to RF; [4C(1)] “Since the feature maps computed at this layer are smaller than the original image, we up-sampled the feature maps using bilinear interpolation to obtain the feature maps FM′(x,y) with the same size as that of the input image I(x,y)”), wherein output of a final layer of the deep learning model is not input to the additional component (Fig. 4; RF receives hidden layer output of U-NET, but not output of a final layer, noted as “Detection by Softmax”, wherein the softmax is also characterized as being of the U-NET model; [4C(1)] “Softmax After fine-tuning was complete, the softmax of the U-Net model was used to produce a pixel-level prediction”) PNG media_image2.png 425 867 media_image2.png Greyscale With respect to claim 2, Dong discloses: The system of claim 1, wherein the information determined for the specimen by the deep learning model is not input to the additional component (Fig. 4, “Softmax”) With respect to claim 6, Dong discloses: The system of claim 1, wherein the information and the additional information comprise classifications of defects detected on the specimen (Fig. 4, “Probsm(x,y)”, “ProbRF(x,y)”, “FP Removal” (as applied to both softmax and RF); [4C(1) Softmax] “ Given an image I(x,y), the softmax maps the output of the last convolutional layer of the U-Net to a probability distribution ProbSM(x,y) over predicted classes (abnormal or normal); [4C(1) Random Forest] “This set was used to train an RF classifier. A new image was processed by applying the CNN encoder, extracting features at each pixel and applying the RF classifier. The result is a probability map”; [4C(3) False Positive Removal] “The final result was a binary mask in which positive pixels indicated the defect/abnormal area”; Fig. 6) With respect to claim 7, Dong discloses: The system of claim 1, wherein the information and the additional information comprise classifications of defects on the specimen as defects of interest or nuisances (Fig. 4, “FP Removal” used for both random forest and U-Net; [4C(3) False Positive Removal] “The high-level defect refinement used by Dong et al. [4] was used to remove false positives”) With respect to claim 8, Dong discloses: The system of claim 1, wherein the information and the additional information are a same type of information (Fig. 4, “Probsm(x,y)”, “ProbRF(x,y)”, “FP Removal” (as applied to both softmax and RF); [4C(1) Softmax] “ Given an image I(x,y), the softmax maps the output of the last convolutional layer of the U-Net to a probability distribution ProbSM(x,y) over predicted classes (abnormal or normal); [4C(1) Random Forest] “This set was used to train an RF classifier. A new image was processed by applying the CNN encoder, extracting features at each pixel and applying the RF classifier. The result is a probability map”; [4C(3) False Positive Removal] “The final result was a binary mask in which positive pixels indicated the defect/abnormal area”; Fig. 6) With respect to claim 10, Dong discloses: The system of claim 1, wherein the deep learning model is further configured as a deep learning neural network (Fig. 4, “U-Net”) With respect to claim 11, Dong discloses: The system of claim 1, wherein the additional component comprises a random forest decision tree (Fig. 4, “Random Forest”) With respect to claim 12, Dong discloses: The system of claim 1, wherein the additional component is further configured as a non-deep learning model (Fig. 4, “Random Forest”) With respect to claim 13, Dong discloses: The system of claim 1, wherein the input specific to the specimen comprises information for a region on the specimen at which the output was generated by the one or more detectors (Fig. 4; [4C(1)] “Since the feature maps computed at this layer are smaller than the original image, we up-sampled the feature maps using bilinear interpolation to obtain the feature maps FM′(x,y) with the same size as that of the input image I(x,y)) With respect to claim 17, Dong discloses: The system of claim 1, wherein the hidden layer output generated by the at least one of the hidden layers comprises multiple different results (Fig. 4, U-Net multiple pooling and convolution stages), wherein the input specific to the specimen comprises multiple different inputs (Fig. 4, “Interpolation”, noting that interpolation generates multiple different scalar values), and wherein the additional component is further configured for independently applying an importance to at least one of the multiple different results and the multiple different inputs prior to determining the additional information (Fig. 4, Random Forest voting ingesting at least one hidden layer result in addition to interpolated feature values) With respect to claim 18, Dong discloses: The system of claim 1, further comprising the output generation subsystem, wherein the one or more detectors are configured for generating the output by detecting light or electrons from the specimen ([6A(1)] “This dataset was acquired by Dong et al. [4] in the factory of an aerospace manufacturing company, where a digital radiography device designed by Bosello was used to acquire the images”) With respect to claim 19, the system of Dong further discloses the recited computer-implementation ([4A] “In all experiments, an NVIDIA Titan XP graphics card was used together with a Lenovo P520 workstation”; wherein a standard Lenovo P520 workstation inherently possesses non-transitory memory for storing instructions). Accordingly, claim 19 (as mirror to claim 1) is rejected in line with the analysis above. With respect to claim 20, the system of claim 1 is configured to perform the method of claim 20. Accordingly, claim 20 is rejected in line with the analysis presented above. 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. Claims 3 is rejected under 35 U.S.C. 103 as being unpatentable over Dong. With respect to claim 3, Dong teaches: The system of claim 1, wherein the additional component is further configured so that the output of the one or more detectors cannot be input to the additional component (Fig. 4; it is not explicitly stated that the original image cannot be input directly to the RF, but any difference would represent no more than an obvious distinction, for the advantage of reducing computational redundancy. See MPEP § 2144.04(V)(C) – “Making Separable”) Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Dong in view of Rahman et. al U-net based defects inspection in photovoltaic electroluminecscence images (Hereinafter, “Rahman”). With respect to claim 4, Dong teaches the system of claim 1 Dong does not explicitly teach the further limitations of claim 4 However, Rahman, in the same field of endeavor of industrial inspection, teaches: wherein at least one of the hidden layers comprises a final fully connected layer in the deep learning model (Fig. 3, “Fully Connected”; Fig. 4; [3A]) It would have been obvious to a person of ordinary skill in the art as of the effective filing date of the claimed invention to modify Dong to include the limitation of a final fully connected layer in the deep learning model, as taught by Rahman. Doing so would allow the network to weigh the channels “based on their importance and allocate high weights to the most important features” (Rahman, [3A]). The systems readily integrate, as Rahman directly evidences that fully connected layers were known to be compatible with U-Net architecture as of the effective filing date of the claimed invention. Claim 5 is rejected under 35 U.S.C. 103 as being unpatentable over Dong in view of Truong et. al Unsupervised Detection of Anomalous Sound for Machine Condition Monitoring using Fully Connected U-Net (Hereinafter, “Truong”) With respect to claim 5, Dong teaches: The system of claim 1, wherein the deep learning model is further configured for determining the information by distillation of the output generated by the at least one of the one or more detectors to a number of pertinent features of the output, and wherein the hidden layer output input to the additional component comprises (Fig. 4; U-Net downsampling) Dong does not explicitly teach: activations of a final fully connected layer in the deep learning model responsive to the distillation However, Troung, in the same field of endeavor of industrial inspection, teaches: activations of a final fully connected layer in the deep learning model responsive to the distillation (Fig. 5; [2.2] “Our new proposed architecture, a fully connected architecture based on the original U-Net called Fully Connected U Net, solves this problem by replacing convolution layers with dense/fully connected layers”) It would have been obvious to a person of ordinary skill in the art as of the effective filing date of the claimed invention to include the limitations of fully-connected U-Net architecture, as taught by Truong. Doing so “inherits the benefits of keeping information from the encoder to the decoder to minimize the reconstruction error in normal data”, (Truong, [2.2]). The systems readily integrate. As combined with Dong, the encoder could be modified to include fully-connected layers. Truong includes further fully-connected layers past the encoder stage (and going into the decoder), but a person of ordinary skill in the art “is a person of ordinary creativity” (See MPEP § 2141.03(I)) and need not include the entirety of a secondary reference’s teachings when making modifications. In fact, Truong directly attributes the success of its modified U-Net to its encoder fully connected layers specifically ([2.2] “In short, Fully Connected U-Net performs better because this architecture uses a fully connected layer instead of the convolutional layer in the original U-Net”). It is also noted that Truong operates on 1D data, however the distinction is also without non-obvious effect. U-Nets are well-established for considering both 1D and 2D data, as evidenced by the body of prior art cited by both Dong and Truong alike. The same advantages advanced in the processing of 1D data are also applicable to the processing of 2D data (retaining information). See MPEP § 2143(I)(F) – “Known work in one field of endeavor may prompt variations of it for use in either the same field or a different one based on design incentives or other market forces if the variations are predictable to one of ordinary skill in the art”. Claim 9 is rejected under 35 U.S.C. 103 as being unpatentable over Dong in view of He et. al Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition (Hereinafter, “He”). With respect to claim 9, Dong teaches: The system of claim 1, wherein the information and the additional information are different kinds of information (in certain embodiments of Dong (Table 6), the SVM of Fig. 4 does not receive prediction mask output and thus can be almost mapped to the “additional component”. However, because it does not receive “input specific”, it does not fulfill claim 1, and therefore cannot anticipate claim 9 either) Dong does not explicitly teach: [input specific to the specimen being combined with feature information with respect to the additional SVM classifier] However, He, in the same field of endeavor of image analysis, teaches: [input specific to the specimen being combined with feature information with respect to the additional SVM classifier] (Fig. 3; [2.2] “Spatial pyramid pooling improves BoW in that it can maintain spatial information by pooling in local spatial bins. These spatial bins have sizes proportional to the image size, so the number of bins is fixed regardless of the image size…With spatial pyramid pooling, the input image can be of any sizes. This not only allows arbitrary aspect ratios, but also allows arbitrary scales”) It would have been obvious to a person of ordinary skill in the art as of the effective filing date of the claimed invention to modify Dong to include the limitations of specimen-specific scaling, as taught by He. Doing so avoids image stretching or squishing, allowing the SVM to accurately classify defects on specimen images of any size or shape. The systems readily integrate, as the additional pre-processing could be performed prior to the ingestion by the SVM, in a comparable fashion to the interpolation pre-processing of the RF. The method of He is also cited by Dong ([4D(1)]), with its spatial pooling explicitly incorporated at least in part already ([4D(1)]). Claims 14-16 are rejected under 35 U.S.C. 103 as being unpatentable over Dong in view of Kwon et. al Fast defect detection for various types of surfaces using random forest with VOV features (Hereinafter, “Kwon”) With respect to claim 14, Dong teaches the system of claim 1 Dong does not explicitly teach the further limitations of claim 14 However, Kwon, in the same field of endeavor of defect detection, teaches: wherein the input specific to the specimen comprises information generated by a defect detection algorithm ([2.2] “The definition of VOV and its calculation process are given in Fig. 2”) applied to the output generated by the at least one of the one or more detectors ([2.1] “Though simple, VOV takes the role of amplifying variation of intensity. With VOV, it is possible to find defects through irregular changes different from the background”; [3] “Several features can be created from VOV values using different window sizes and directions. Using a feature vector f created in this way, classifiers in a classification method can be trained to recognize defects…The feature vector consists of two basic feature groups such as 16 original VOV features according to size and direction of defect and 6 application features of VOV”; equations 8-11) PNG media_image3.png 792 582 media_image3.png Greyscale It would have been obvious to a person of ordinary skill in the art as of the effective filing date of the claimed invention to include the limitations of additional defect detection algorithm information, as taught by Kwon. Doing so would have the advantage of supplying additional weighting information to the random forest classifier, improving accuracy. The systems readily integrate, as Kwon directly evidences that VOV-derived features are compatible with Random Forest, which is also used by Dong as the “additional component”. With respect to claim 15, Dong teaches the system of claim 1 Dong does not explicitly teach the further limitations of claim 15 However, Kwon teaches: wherein the input specific to the specimen comprises information specific to at least one defect on the specimen generated by a detection algorithm ([2.2] “The definition of VOV and its calculation process are given in Fig. 2”; [3] “The feature vector consists of two basic feature groups such as 16 original VOV features according to size and direction of defect and 6 application features of VOV”) applied to the output generated by the at least one of the one or more detectors ([2.1] “Though simple, VOV takes the role of amplifying variation of intensity. With VOV, it is possible to find defects through irregular changes different from the background”; [3] “Several features can be created from VOV values using different window sizes and directions. Using a feature vector f created in this way, classifiers in a classification method can be trained to recognize defects…The feature vector consists of two basic feature groups such as 16 original VOV features according to size and direction of defect and 6 application features of VOV”; equations 8-11) It would have been obvious to a person of ordinary skill in the art as of the effective filing date of the claimed invention to include the limitations of additional defect detection algorithm information, as taught by Kwon. Doing so would have the advantage of supplying additional weighting information to the random forest classifier, improving accuracy. The systems readily integrate, as Kwon directly evidences that VOV-derived features are compatible with Random Forest, which is also used by Dong as the “additional component”. With respect to claim 16, Dong teaches the system of claim 1 Dong does not explicitly teach the further limitations of claim 16 However, Kwon teaches: wherein the input specific to the specimen comprises a parameter of a defect detection algorithm ([2.2] “The definition of VOV and its calculation process are given in Fig. 2”; [3] “…VOV parameter…”) applied to the output generated by the at least one of the one or more detectors ([2.1] “Though simple, VOV takes the role of amplifying variation of intensity. With VOV, it is possible to find defects through irregular changes different from the background”; [3] “Several features can be created from VOV values using different window sizes and directions. Using a feature vector f created in this way, classifiers in a classification method can be trained to recognize defects…The feature vector consists of two basic feature groups such as 16 original VOV features according to size and direction of defect and 6 application features of VOV”; equations 8-11) It would have been obvious to a person of ordinary skill in the art as of the effective filing date of the claimed invention to include the limitations of additional defect detection algorithm information, as taught by Kwon. Doing so would have the advantage of supplying additional weighting information to the random forest classifier, improving accuracy. The systems readily integrate, as Kwon directly evidences that VOV-derived features are compatible with Random Forest, which is also used by Dong as the “additional component”. Inquiry Any inquiry concerning this communication or earlier communications from the examiner should be directed to NOAH WILLIAM BOYAR whose telephone number is 571-272- 8392. The examiner can normally be reached 10:00 AM – 6:00 PM EST, Monday – Friday. 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 https://www.uspto.gov/interviewpractice. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Chan Park can be reached at 571-272-7409. 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. /NOAH W BOYAR/ Examiner, Art Unit 2669 /CHAN S PARK/Supervisory Patent Examiner, Art Unit 2669
Read full office action

Prosecution Timeline

Jun 05, 2024
Application Filed
Mar 24, 2026
Non-Final Rejection mailed — §102, §103, §112
Jun 15, 2026
Response Filed
Jul 02, 2026
Final Rejection mailed — §102, §103, §112
Sep 01, 2026
Response after Non-Final Action
Sep 27, 2026
Request for Continued Examination
Oct 05, 2026
Response after Non-Final Action
Nov 12, 2026
Non-Final Rejection mailed — §102, §103, §112 (current)

Precedent Cases

Applications granted by this same examiner with similar technology

Patent 12802162
PADDING METHOD, APPARATUS, SYSTEM AND COMPUTER READABLE STORAGE MEDIUM
4y 0m to grant Granted Nov 17, 2026
Patent 12770279
ESTIMATION DEVICE
2y 7m to grant Granted Oct 20, 2026
Patent 12770266
SINGLE VERTEBRA SEGMENTATION METHOD, DEVICE, EQUIPMENT AND STORAGE MEDIUM
2y 4m to grant Granted Oct 20, 2026
Study what changed to get past this examiner. Based on 3 most recent grants.

Strategy Recommendation AI-generated — please review before filing

Get a prosecution strategy drawn from examiner precedents, rejection analysis, and claim mapping.
Typically takes 5-10 seconds — AI-generated, attorney review required before filing

Prosecution Projections

3-4
Expected OA Rounds
100%
Grant Probability
99%
With Interview (+0.0%)
2y 6m (~0m remaining)
Median Time to Grant
High
PTA Risk
Based on 10 resolved cases by this examiner. Grant probability derived from career allowance rate.

Sign in with your work email

Enter your email to receive a magic link. No password needed.

Personal email addresses (Gmail, Yahoo, etc.) are not accepted.

Free tier: 3 strategy analyses per month