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 .
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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 10 September 2026 has been entered.
Response to Amendment
This action is responsive to the request for continued examination (RCE) received 10 September 2026 and the amendments and remarks received 19 August 2026. Claims 1 - 20 are currently pending.
The amendment to the claims filed on 19 August 2026 does not comply with the requirements of 37 CFR 1.121(c) because claim 14 has not been provided with the proper status identifier and because claims 4, 11, 14 and 18 do not include proper markings to indicate the changes that have been made relative to the immediate prior version of the claims, for example, see line 5 and line 11 of claim 4 and line 2 of claim 14. Amendments to the claims filed on or after July 30, 2003 must comply with 37 CFR 1.121(c) which states:
(c) Claims. Amendments to a claim must be made by rewriting the entire claim with all changes (e.g., additions and deletions) as indicated in this subsection, except when the claim is being canceled. Each amendment document that includes a change to an existing claim, cancellation of an existing claim or addition of a new claim, must include a complete listing of all claims ever presented, including the text of all pending and withdrawn claims, in the application. The claim listing, including the text of the claims, in the amendment document will serve to replace all prior versions of the claims, in the application. In the claim listing, the status of every claim must be indicated after its claim number by using one of the following identifiers in a parenthetical expression: (Original), (Currently amended), (Canceled), (Withdrawn), (Previously presented), (New), and (Not entered).
(1) Claim listing. All of the claims presented in a claim listing shall be presented in ascending numerical order. Consecutive claims having the same status of “canceled” or “not entered” may be aggregated into one statement (e.g., Claims 1–5 (canceled)). The claim listing shall commence on a separate sheet of the amendment document and the sheet(s) that contain the text of any part of the claims shall not contain any other part of the amendment.
(2) When claim text with markings is required. All claims being currently amended in an amendment paper shall be presented in the claim listing, indicate a status of “currently amended,” and be submitted with markings to indicate the changes that have been made relative to the immediate prior version of the claims. The text of any added subject matter must be shown by underlining the added text. The text of any deleted matter must be shown by strike-through except that double brackets placed before and after the deleted characters may be used to show deletion of five or fewer consecutive characters. The text of any deleted subject matter must be shown by being placed within double brackets if strike-through cannot be easily perceived. Only claims having the status of “currently amended,” or “withdrawn” if also being amended, shall include markings. If a withdrawn claim is currently amended, its status in the claim listing may be identified as “withdrawn—currently amended.”
(3) When claim text in clean version is required. The text of all pending claims not being currently amended shall be presented in the claim listing in clean version, i.e., without any markings in the presentation of text. The presentation of a clean version of any claim having the status of “original,” “withdrawn” or “previously presented” will constitute an assertion that it has not been changed relative to the immediate prior version, except to omit markings that may have been present in the immediate prior version of the claims of the status of “withdrawn” or “previously presented.” Any claim added by amendment must be indicated with the status of “new” and presented in clean version, i.e., without any underlining.
(4) When claim text shall not be presented; canceling a claim.
(i) No claim text shall be presented for any claim in the claim listing with the status of “canceled” or “not entered.”
(ii) Cancellation of a claim shall be effected by an instruction to cancel a particular claim number. Identifying the status of a claim in the claim listing as “canceled” will constitute an instruction to cancel the claim.
(5) Reinstatement of previously canceled claim. A claim which was previously canceled may be reinstated only by adding the claim as a “new” claim with a new claim number.
Claim Objections
Claim 4 is objected to because of the following informalities: Lines 1 - 7 of claim 4 recite, in part, “method of claim 1, further comprises: determining,” which appears to contain a grammatical error and/or a minor informality. The Examiner suggests amending the claim to --method of claim 1, further comprising: determining,-- in order to improve the clarity and precision of the claim. Appropriate correction is required.
Claim 4 is objected to because of the following informalities: Lines 8 - 10 of claim 4 recite, in part, “the at least one WBC and the at least one RBC in the at least one training image; cropping, by the processor,” which appears to contain inconsistent claim terminology, a grammatical error and/or minor informalities. The Examiner suggests amending the claim to --the at least one WBC [[and]] and/or the at least one RBC in the at least one training image; and cropping, by the processor,-- in order to maintain consistency with lines 18 - 19 of claim 1 and to improve the clarity and precision of the claim. Appropriate correction is required.
Claim 4 is objected to because of the following informalities: Line 11 of claim 4 recites, in part, “at least one RBC image;” which appears to contain a grammatical error and/or a minor informality. The Examiner suggests amending the claim to --at least one RBC image[[;]]-- in order to improve the clarity and precision of the claim. Appropriate correction is required.
Claim 5 is objected to because of the following informalities: Lines 1 - 2 of claim 5 recite, in part, “method of claim 1, comprises: inputting,” which appears to contain a grammatical error and/or a minor informality. The Examiner suggests amending the claim to --method of claim 1, further comprising: inputting,-- in order to improve the clarity and precision of the claim. Appropriate correction is required.
Claim 11 is objected to because of the following informalities: Lines 1 - 7 of claim 11 recite, in part, “wherein the processor is further configured to: determining bounding boxes” which appears to contain a grammatical error and/or a minor informality. The Examiner suggests amending the claim to --wherein the processor is further configured to: determine bounding boxes-- in order to improve the clarity and precision of the claim. Appropriate correction is required.
Claim 11 is objected to because of the following informalities: Lines 8 - 10 of claim 11 recite, in part, “the at least one WBC and the at least one RBC in the at least one training image; cropping each of the bounding boxes” which appears to contain inconsistent claim terminology, grammatical errors and/or minor informalities. The Examiner suggests amending the claim to --the at least one WBC [[and]] and/or the at least one RBC in the at least one training image; and crop each of the bounding boxes-- in order to maintain consistency with lines 19 - 20 of claim 8 and to improve the clarity and precision of the claim. Appropriate correction is required.
Claim 11 is objected to because of the following informalities: Line 11 of claim 11 recites, in part, “at least one RBC image;” which appears to contain a grammatical error and/or a minor informality. The Examiner suggests amending the claim to --at least one RBC image[[;]]-- in order to improve the clarity and precision of the claim. Appropriate correction is required.
Claim 14 is objected to because of the following informalities: Lines 1 - 2 of claim 14 recite, in part, “wherein the processor is configured to: Classify each of the plurality of blood-cells” which appears to contain a grammatical error and/or a minor informality. The Examiner suggests amending the claim to --wherein the processor is configured to: classify each of the plurality of blood-cells-- in order to improve the clarity and precision of the claim. Appropriate correction is required.
Claim 16 is objected to because of the following informalities: Lines 1 - 2 of claim 16 recite, in part, “wherein the computer executable instructions are configured” which appears to contain inconsistent claim terminology and/or a minor informality. The Examiner suggests amending the claim to --wherein the computer-executable instructions are configured-- in order to maintain consistency with lines 1 - 2 of claim 15 and to improve the clarity and precision of the claim. Appropriate correction is required.
Claim 18 is objected to because of the following informalities: Lines 2 - 8 of claim 18 recite, in part, “the computer executable instructions are configured to: determining bounding boxes for each of the” which appears to contain inconsistent claim terminology, a grammatical error and/or minor informalities. The Examiner suggests amending the claim to --the computer-executable instructions are configured to: determine bounding boxes for each of the-- in order to maintain consistency with lines 1 - 2 of claim 15 and to improve the clarity and precision of the claim. Appropriate correction is required.
Claim 18 is objected to because of the following informalities: Lines 9 - 11 of claim 18 recite, in part, “the at least one WBC and the at least one RBC in the at least one training image; cropping each of the bounding boxes” which appears to contain grammatical errors, inconsistent claim terminology and/or minor informalities. The Examiner suggests amending the claim to --the at least one WBC [[and]] and/or the at least one RBC in the at least one training image; crop each of the bounding boxes-- in order to maintain consistency with lines 19 - 20 of claim 15 and to improve the clarity and precision of the claim. Appropriate correction is required.
Claim 18 is objected to because of the following informalities: Line 12 of claim 18 recites, in part, “at least one RBC image;” which appears to contain a grammatical error and/or a minor informality. The Examiner suggests amending the claim to --at least one RBC image[[;]]-- in order to improve the clarity and precision of the claim. Appropriate correction is required.
Claim 19 is objected to because of the following informalities: Lines 1 - 3 of claim 19 recite, in part, “wherein the computer executable instructions for: inputting the report” which appears to contain inconsistent claim terminology, a grammatical error and/or minor informalities. The Examiner suggests amending the claim to --wherein the computer-executable instructions are further configured for: inputting the report-- in order to maintain consistency with lines 19 - 20 of claim 15 and to improve the clarity and precision of the claim. Appropriate correction is required.
Claim Rejections - 35 USC § 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 1 - 20 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.
Claim 1 recites the limitation "the at least one WBC image" in line 24. There is insufficient antecedent basis for this limitation in the claim.
Claim 1 recites the limitation "the at least one RBC image" in lines 24 - 25. There is insufficient antecedent basis for this limitation in the claim.
Claim 8 recites the limitation "the at least one WBC image" in line 23. There is insufficient antecedent basis for this limitation in the claim.
Claim 8 recites the limitation "the at least one RBC image" in line 24. There is insufficient antecedent basis for this limitation in the claim.
Claim 15 recites the limitation "the at least one WBC image" in lines 22 - 23. There is insufficient antecedent basis for this limitation in the claim.
Claim 15 recites the limitation "the at least one RBC image" in line 23. There is insufficient antecedent basis for this limitation in the claim.
Claims 2 - 7, 9 - 14 and 16 - 20 are also rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, due to being dependent upon a rejected base claim(s) but would be withdrawn from the rejection if their base claim(s) overcome the rejection.
Response to Arguments
Applicant's arguments filed 19 August 2026 have been fully considered but they are not persuasive.
On pages 16 - 22 of the remarks the Applicant’s Representative argues that the previously cited references, alone or in combination, “fail to teach, disclose, or suggest ‘...inputting at least one training image to the deep learning mode, [sic] wherein the at least one training image comprises at least one WBC and/or at least one RBC; determining the plurality of images of WBCs and RBCs by: generating a first set of samples corresponding to WBCs and a second set of samples corresponding to RBCs based on the at least one WBC image and the at least one RBC image respectively using a generator model of the deep learning model, wherein the first set of samples and the second set of samples are generated based on a classification of each sample as one of real sample or fake sample using a discriminator model of the deep learning model, ... wherein the plurality of images of WBCs and RBCs are determined based on the first set of samples and the second set of samples, respectively’ as recited in independent claims 1, 8, and 15.”
In particular, on pages 19 - 20 of the remarks, the Applicant’s Representative argues that Soni et al. do not teach or suggest “using the GAN discriminator itself as the claimed realism-validation mechanism” at least because Soni et al. disclose using “a separate DNN for classifying synthetic images as real-like or non-real-like.” Therefore, the Applicant’s Representative argues that Soni et al. do not teach or suggest “using the GAN discriminator to determine whether the generated WBC and RBC samples satisfy the claimed real-sample condition, much less using that determination together with a WBC/RBC balancing condition to control continued sample generation” and that the previously cited references alone or in combination fail to teach, disclose or suggest the aforementioned disputed claim limitation(s).
The Examiner respectfully disagrees.
In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986).
In response to applicant's argument that the references fail to show certain features of the invention, it is noted that the features upon which applicant relies (i.e., “using the GAN discriminator itself (emphasis added) as the claimed realism-validation mechanism”) are not recited in the rejected claim(s). Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993). The Examiner asserts that the claim limitations of instant claims 1, 8 and 15 corresponding to the aforementioned disputed feature(s) recite and require “wherein the first set of samples and the second set of samples are generated based on a classification of each sample as one of real sample or fake sample using a discriminator model of the deep learning model”.
Furthermore, the Examiner asserts that Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. disclose the aforementioned disputed claim limitation(s), see at least the analysis of claims 1, 8 and 15 included herein below in section 30 of the instant Office Action.
The Examiner asserts that Tandon et al. disclose obtaining a first set of samples corresponding to WBCs and a second set of samples corresponding to RBCs, see at least page 12 paragraphs 0212 - 0219, page 13 paragraph 0221, page 14 paragraphs 0253 - 0259 and page 16 paragraphs 0289 - 0296 of Tandon et al., and that Jung et al. disclose using a GAN discriminator to determine whether generated samples satisfy the claimed real-sample condition and generating first and second sets of samples until the first set is balanced with respect to the second set, see at least page 1 abstract, page 2 subsection “Contributions”, page 10 subsection “Dataset sharing” - page 12 subsection “Conclusion” and page 11 figure 4 of Jung et al. wherein they disclose that they “generate and publicize synthetic WBC images using Generative Adversarial Network to be used for education and research purposes. The synthetic WBC images are verified by experiments and a domain expert to have a high degree of similarity to the original images”, that “GAN is a deep learning architecture for generating new artificial samples, it composes of two deep networks: ❶ the generator G, and ❷ the discriminator D. The G generates new samples from the domain, and the D classifies whether the samples are real or fake”, that they train a “Deep Convolutional Generative Adversarial Network (DCGAN) [70] to train (G and D) models for generating images” and that they “generated 1000 plausible WBC images of size of 128x128x3 for each class (a total of 5000 images).”
The Examiner asserts that, as shown herein above and in the cited portions, Tandon et al. disclose obtaining and utilizing a training set of WBC samples and RBC samples, and that Jung et al. disclose generating new artificial image samples using a discriminator of a GAN that classifies whether the samples are real or fake and until an equal number of plausible samples are generated for each of a plurality of blood cell classes.
Moreover, the Examiner asserts that Soni et al. disclose using a discriminator model of a deep learning model to determine whether generated samples satisfy the claimed real-sample condition, see at least figures 1, 3 - 6, 9 and 10, page 2 paragraph 0019, page 3 paragraphs 0026 and 0030, page 4 paragraphs 0036 and 0038 - 0041, page 5 paragraphs 0043 and 0047 and page 6 paragraph 0052 of Soni et al. wherein they disclose “automatically identifying and extracting realistic synthetic images as opposed to non-realistic synthetic images using a deep neural network (DNN) model trained on real images and previously identified synthetic images classified as non-real-like. The synthetic images can include images generated using a generative adversarial network (GAN) model. The identified and extracted realistic synthetic images can be combined with real images to form a comprehensive training dataset that can be used to train a target machine learning model to perform various automated image analysis functions. For example, in various embodiments, synthetic images can include synthetic medical images generated using a GAN model” [0019], that “once the DNN model 126 training is complete, the DNN model 126 can be used to automatically classify synthetic images generated using the GAN model 108 as either real-like or non-real-like” [0041] and that “after the DNN model 126 has been trained, the synthetic image generation module 102 can generate second synthetic images using the synthetic image generator component 106, the GAN model 108 and the latent random variable data 104. The inference component 406 can further employ the DNN model 126 to classify each (or in some implementations one or more) of the second synthetic images as either real-like or non-real-like. The second extraction component 408 can further extract all (or in some implementations one or more) of the second synthetic images classified as real-like, depicted in FIG. 4 as real-like synthetic images 410. These real-like synthetic images can further be added to the target model dataset 118. In some embodiments, the second extraction component 408 can discard the second synthetic images 402 classified as non-real-like” [0047].
The Examiner asserts that, as shown herein above and in the cited portions, Soni et al. disclose that only synthetic medical images that are classified as real-like by a trained DNN model, a discriminator of the deep learning model, are added to a target model dataset for subsequent uses as training images.
Therefore, the Examiner asserts that Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. disclose the aforementioned disputed claim limitation(s).
On pages 16 - 22 of the remarks the Applicant’s Representative argues that the previously cited references, alone or in combination, “fail to teach, disclose, or suggest ‘...inputting at least one training image to the deep learning mode, [sic] wherein the at least one training image comprises at least one WBC and/or at least one RBC; determining the plurality of images of WBCs and RBCs by: generating a first set of samples corresponding to WBCs and a second set of samples corresponding to RBCs based on the at least one WBC image and the at least one RBC image respectively using a generator model of the deep learning model, wherein the first set of samples and the second set of samples are generated based on a classification of each sample as one of real sample or fake sample using a discriminator model of the deep learning model, ... wherein the plurality of images of WBCs and RBCs are determined based on the first set of samples and the second set of samples, respectively’ as recited in independent claims 1, 8, and 15.”
In particular, on page 19 of the remarks, the Applicant’s Representative argues that Jung et al. do not disclose “generating a first set of samples corresponding to WBCs and a second set of samples corresponding to RBCs and continuing generation until the first set is balanced with respect to the second set.”
The Examiner respectfully disagrees.
In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986).
Furthermore, the Examiner asserts that Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. disclose the aforementioned disputed claim limitation(s), see at least the analysis of claims 1, 8 and 15 included herein below in section 30 of the instant Office Action.
The Examiner asserts that Tandon et al. disclose obtaining a first set of samples corresponding to WBCs and a second set of samples corresponding to RBCs, see at least page 12 paragraphs 0212 - 0219, page 13 paragraph 0221, page 14 paragraphs 0253 - 0259 and page 16 paragraphs 0289 - 0296 of Tandon et al., and that Jung et al. disclose generating first and second sets of samples until the first set is balanced with respect to the second set, see at least page 2 subsection “Contributions” and page 10 subsection “Dataset sharing” - page 12 subsection “Conclusion” of Jung et al. wherein they disclose that they train a “Deep Convolutional Generative Adversarial Network (DCGAN) [70] to train (G and D) models for generating images” and that they “generated 1000 plausible WBC images of size of 128x128x3 for each class (a total of 5000 images).”
The Examiner asserts that, as shown herein above and in the cited portions, Tandon et al. disclose obtaining and utilizing a training set of WBC samples and RBC samples, and that Jung et al. disclose generating new artificial image samples using a GAN until an equal number of plausible samples are generated for each of a plurality of blood cell classes. The Examiner asserts that in the proposed combination the teachings of Tandon et al. would be modified by the teachings of Jung et al. such that a training set comprising a first set of WBC samples and a second set of RBC samples would be generated such that the first set of WBC samples is balanced with respect to the second set of RBC samples.
Therefore, the Examiner asserts that Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. disclose the aforementioned disputed claim limitation(s).
On pages 16 - 22 of the remarks the Applicant’s Representative argues that the previously cited references, alone or in combination, “fail to teach, disclose, or suggest ‘...inputting at least one training image to the deep learning mode, [sic] wherein the at least one training image comprises at least one WBC and/or at least one RBC; determining the plurality of images of WBCs and RBCs by: generating a first set of samples corresponding to WBCs and a second set of samples corresponding to RBCs based on the at least one WBC image and the at least one RBC image respectively using a generator model of the deep learning model, wherein the first set of samples and the second set of samples are generated based on a classification of each sample as one of real sample or fake sample using a discriminator model of the deep learning model, ... wherein the plurality of images of WBCs and RBCs are determined based on the first set of samples and the second set of samples, respectively’ as recited in independent claims 1, 8, and 15.”
In particular, on page 20 of the remarks, the Applicant’s Representative argues that Jung et al. do not disclose “the claimed contour-based determination of individual WBC and RBC bounding boxes followed by cropping of those bounding boxes to obtain the WBC and RBC images used for the claimed sample-generation process.”
The Examiner respectfully disagrees.
In response to applicant's arguments against the references individually, one cannot show nonobviousness by attacking references individually where the rejections are based on combinations of references. See In re Keller, 642 F.2d 413, 208 USPQ 871 (CCPA 1981); In re Merck & Co., 800 F.2d 1091, 231 USPQ 375 (Fed. Cir. 1986).
Furthermore, the Examiner asserts that Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. disclose the aforementioned disputed claim limitation(s), see at least the analysis of claims 1, 4, 8, 11, 15 and 18 included herein below in section 30 of the instant Office Action.
The Examiner asserts that, at least, Tandon et al. disclose the aforementioned disputed claim limitation(s), see at least figures 23 - 25, page 6 paragraphs 0146 - 0147, page 12 paragraphs 0212 - 0219, page 14 paragraphs 0262 - 0263, page 16 paragraphs 0285 - 0290 and page 18 paragraphs 0330 - 0334 of Tandon et al. wherein they disclose that “a cellular artifact represents a collection of contiguous pixels (with associated position and magnitude values) that are identified as likely belonging to a cell, parasite, or other sample feature of interest in a biological sample. Typically, the collection of contiguous pixels is within or proximate to a boundary defined through segmentation. Often, a cellular artifact includes pixels of an identified boundary, all pixels within that boundary, and optionally some relatively small number of pixels surrounding the boundary” [0147], that “the image is then spliced based on the derived artifact dimensions from Sobel filtering elevation map generation given by equation 4. The Elevation map technique uses a Sobel operator to approximate the size of each artifact and conduct the cell extraction accordingly” [0286], that the “spliced image is generated through the numpy sub array function passing the Euclidean and Sobel rectangular values as parameters, resulting in a dataset of segmented cells from the original smear image shot. The coordinates extracted from the Euclidean distance transformation and local maxima calculation are applied back to the original colored image to make the rectangular segmentation of cells on the original. The cells are then normalized to a 50×50 jpeg shot for identification” [0289] and that “segmentation further involves splicing the one or more images of the biological sample using the local maxima and data obtained from applying the Sobel filter, thereby obtaining a plurality of images of the cellular artifacts. In some applications, each spliced image includes a cellular artifact” [0332].
The Examiner asserts that, as shown herein above and in the cited portions, Tandon et al. disclose that a cellular artifact represents a collection of contiguous pixels that are identified as likely belonging to a cell, that the collection of contiguous pixels is within or proximate to a boundary defined through segmentation, that a Sobel operator is used to approximate the size of each artifact and conduct the cell extraction, that a spliced image is generated through the numpy sub array function passing the Euclidean and Sobel rectangular values as parameters, resulting in a dataset of segmented cells from the original smear image shot, and that their segmentation further involves splicing one or more images of the biological sample using the local maxima and data obtained from applying a Sobel filter to obtain a plurality of images of cellular artifacts. Thus, the Examiner asserts that, at least, Tandon et al. disclose the aforementioned disputed claim limitation at least because Tandon et al. disclose identifying collections of contiguous pixels that are within or proximate to boundaries defined through segmentation, i.e., bounding boxes are determined for each blood cell based on contour detected through segmentation, and that their segmentation further involves splicing one or more images of the biological sample to obtain a plurality of images of cellular artifacts, i.e., the bounding boxes determined for each blood cell are cropped.
In addition, the Examiner asserts that Jung et al. disclose cropping each of the bounding boxes to determine at least one WBC image and at least one RBC image, see at least page 5 subsection “Pre-processing of WBC images”, and page 5 figures 1 and 2 of Jung et al. wherein they disclose that “WBC images are pre-processed using three steps: ❶ image border cropping, ❷ image re-sizing, and ❸ image normalization.”
Therefore, the Examiner asserts that Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. disclose the aforementioned disputed claim limitation(s).
On page 21 of the remarks the Applicant’s Representative argues that “arriving at the claimed combination would require modifying the teachings of Tandon, Jung, and Soni in a manner that is not taught or suggested by the references and would amount to reconstruction of the claimed invention with the benefit of hindsight.” Therefore, the Applicant’s Representative argues that “the cited references, individually or in combination, fail to render the claimed invention obvious.”
The Examiner respectfully disagrees.
In response to applicant's argument that the examiner's conclusion of obviousness is based upon improper hindsight reasoning, it must be recognized that any judgment on obviousness is in a sense necessarily a reconstruction based upon hindsight reasoning. But so long as it takes into account only knowledge which was within the level of ordinary skill at the time the claimed invention was made, and does not include knowledge gleaned only from the applicant's disclosure, such a reconstruction is proper. See In re McLaughlin, 443 F.2d 1392, 170 USPQ 209 (CCPA 1971).
Furthermore, the Examiner asserts that Tandon et al., El-Zehiry et al., Jung et al., and Soni et al. suggest and disclose the aforementioned disputed claim limitations, see at least the discussion included herein above in sections 27a - 27c of the instant Office Action.
Therefore, the Examiner asserts that Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. disclose the aforementioned disputed claim limitation(s).
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 text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claims 1, 3, 4, 7, 8, 10, 11, 14, 15, 17 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Tandon et al. U.S. Publication No. 2018/0211380 A1 in view of El-Zehiry et al. U.S. Publication No. 2017/0132450 A1 in view of Jung et al., "WBC image classification and generative models based on convolutional neural network", BMC Medical Imaging, Vol. 22, No. 94, May 2022, pages 1 - 16 in view of Soni et al. U.S. Publication No. 2020/0311913 A1.
- The Examiner notes, with regards to claims 15 - 20, that claims 15 - 20 do not positively recite an interrelationship between the computer-executable instructions and an intended computer system for executing the computer-executable instructions and absent such a positively recited interrelationship the broadest reasonable interpretation of the limitations that the computer-executable instructions are intended to perform encompass interpretations wherein those limitations are non-functional because the claims do not limit the computer-executable instructions to an embodiment wherein the computer-executable instructions are executed by an intended computer system in order to perform its recited limitations.
- With regards to claims 1, 8 and 15, Tandon et al. disclose a method of analysing a blood smear image, (Tandon et al., Figs. 4A - 8, 10, 12, 18, 19, 24 & 25, Pg. 5 ¶ 0093 and 0096, Pg. 6 ¶ 0143 - 0147, Pg. 8 ¶ 0160 - Pg. 9 ¶ 0171, Pg. 10 ¶ 0193 - 0194, Pg. 11 ¶ 0205 - 0207, Pg. 12 ¶ 0210 - 0217) a system of analysing a blood smear image, (Tandon et al., Figs. 4A - 8, 10, 12, 18, 19, 24 & 25, Pg. 5 ¶ 0093 and 0096, Pg. 6 ¶ 0143 - 0147, Pg. 8 ¶ 0160 - Pg. 9 ¶ 0171, Pg. 10 ¶ 0193 - 0194, Pg. 11 ¶ 0205 - 0207, Pg. 12 ¶ 0210 - 0217, Pg. 22 ¶ 0380 - Pg. 23 ¶ 0388) comprising: a processor; (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) and a memory communicably coupled to the processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) wherein the memory stores processor-executable instructions, which, on executing by the processor, cause the processor to [perform operations], (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) and a non-transitory computer-readable medium storing computer-executable instructions for analysing a blood smear image, (Tandon et al., Figs. 4A - 8, 10, 12, 18, 19, 24 & 25, Pg. 3 ¶ 0038, Pg. 5 ¶ 0093 and 0096, Pg. 6 ¶ 0143 - 0147, Pg. 8 ¶ 0160 - Pg. 9 ¶ 0171, Pg. 10 ¶ 0193 - 0194, Pg. 11 ¶ 0205 - 0207, Pg. 12 ¶ 0210 - 0217, Pg. 22 ¶ 0381) the computer-executable instructions configured for: detecting, by a processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) a plurality of blood cells in the blood smear image based on edge detection of the plurality of blood cells, (Tandon et al., Figs. 5, 12, 24 & 25, Pg. 6 ¶ 0143 and 0146 - 0147, Pg. 14 ¶ 0260 - 0263, Pg. 15 ¶ 0273 - 0277, Pg. 16 ¶ 0286 - 0290, Pg. 18 ¶ 0331 - 0332) wherein the edge detection of the plurality of blood cells in the blood smear image is based on a preprocessing of the blood smear image; (Tandon et al., Pg. 14 ¶ 0260 - 0268, Pg. 16 ¶ 0286 - 0290, Pg. 18 ¶ 0330 - 0332) determining, by the processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) contours of each of the plurality of blood cells based on the edge detection of the plurality of blood-cells; (Tandon et al., Pg. 6 ¶ 0146 - 0147, Pg. 14 ¶ 0262 - 0263, Pg. 15 ¶ 0275 - 0276, Pg. 16 ¶ 0286 - 0290, Pg. 18 ¶ 0331 - 0332) determining, by the processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) a bounding box for each of the plurality of blood-cells based on the contours of each of the plurality of blood-cells; (Tandon et al., Figs. 23 - 25, Pg. 6 ¶ 0146 - 0147, Pg. 14 ¶ 0262 - 0263, Pg. 16 ¶ 0285 - 0290, Pg. 18 ¶ 0330 - 0332, 0334 and 0338) classifying, by the processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) each of the plurality of blood-cells as one of a white blood cell (WBC) or a red blood cell (RBC) using a deep learning model, (Tandon et al., Figs. 25, Pg. 7 ¶ 0150 - 0152, Pg. 11 ¶ 0206 - 0207, Pg. 12 ¶ 0210 - 0219, Pg. 13 ¶ 0221 - 0236, Pg. 16 ¶ 0296 - Pg. 17 ¶ 0313, Pg. 18 ¶ 0336) wherein the deep learning model is trained based on training data comprising a plurality of images of WBCs and RBCs; (Tandon et al., Figs. 7 & 8, Pg. 12 ¶ 0210 - 0219, Pg. 13 ¶ 0221 - 0236, Pg. 14 ¶ 0253 - 0256) wherein the training data is generated by: inputting, by the processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) at least one training image to the deep learning model, (Tandon et al., Pg. 12 ¶ 0212 - 0219, Pg. 13 ¶ 0221, Pg. 14 ¶ 0253 - 0259, Pg. 16 ¶ 0289 - 0296) wherein the at least one training image comprises at least one WBC and/or at least one RBC; (Tandon et al., Pg. 12 ¶ 0212 - 0219, Pg. 13 ¶ 0221, Pg. 14 ¶ 0253 - 0259, Pg. 16 ¶ 0289 - 0296) and determining, by the processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) the plurality of images of WBCs and RBCs; (Tandon et al., Pg. 12 ¶ 0212 - 0219, Pg. 13 ¶ 0221, Pg. 14 ¶ 0253 - 0259, Pg. 16 ¶ 0289 - 0296) determining, by the processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) a count of WBCs, (Tandon et al., Pg. 11 ¶ 0205 - 0207, Pg. 12 ¶ 0210, Pg. 21 ¶ 0363 - 0370) and a count of RBCs (Tandon et al., Pg. 11 ¶ 0205 - 0207, Pg. 12 ¶ 0210 - 0211, Pg. 18 ¶ 0336, Pg. 21 ¶ 0365) based on the classification and the contours of each of the plurality of blood-cells; (Tandon et al., Pg. 7 ¶ 0150 - 0154, Pg. 8 ¶ 0160 - 0163, Pg. 11 ¶ 0205 - Pg. 12 ¶ 0211, Pg. 16 ¶ 0285 - Pg. 17 ¶ 0298, Pg. 18 ¶ 0329 - 0336, Pg. 21 ¶ 0363 - 0365) and outputting, by the processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) a report comprising the count of WBC, the count of RBCs or the volumetric information of the RBCs and the WBCs. (Tandon et al., Figs. 4A, 6 & 7, Pg. 7 ¶ 0150 - 0151, Pg. 8 ¶ 0160 - 0165, Pg. 11 ¶ 0205 - Pg. 12 ¶ 0211, Pg. 18 ¶ 0334 - 0336, Pg. 21 ¶ 0362 - 0369) Tandon et al. fail to disclose explicitly generating a first set of samples corresponding to WBCs and a second set of samples corresponding to RBCs based on the at least one WBC image and the at least one RBC image respectively using a generator model of the deep learning model, wherein the first set of samples and the second set of samples are generated based on a classification of each sample as one of real sample or fake sample using a discriminator model of the deep learning model, wherein samples corresponding to the first set of samples and the second set of samples are generated until the first set of samples is balanced with respect to the second set of samples and each of the first set of samples and the second set of samples are classified as real by the discriminator model; and wherein the plurality of images of WBCs and RBCs are determined based on the first set of samples and the second set of samples, respectively; and determining volumetric information of the RBCs and the WBCs. Pertaining to analogous art, El-Zehiry et al. disclose analysing a blood smear image, (El-Zehiry et al., Abstract, Figs. 1, 2, 5 - 7 & 14, Pg. 2 ¶ 0030 - Pg. 3 ¶ 0032, Pg. 3 ¶ 0038 - 0042) comprising: detecting a plurality of blood cells in the blood smear image based on edge detection of the plurality of blood cells, (El-Zehiry et al., Figs. 1, 2, 5 & 6, Pg. 2 ¶ 0031 - Pg. 3 ¶ 0032, Pg. 3 ¶ 0039, Pg. 3 ¶ 0042 - Pg. 4 ¶ 0043) wherein the edge detection of the plurality of blood cells in the blood smear image is based on a preprocessing of the blood smear image; (El-Zehiry et al., Figs. 1, 2, 5 & 6, Pg. 2 ¶ 0031 - Pg. 3 ¶ 0032, Pg. 3 ¶ 0039, Pg. 3 ¶ 0042 - Pg. 4 ¶ 0043) determining contours of each of the plurality of blood cells based on the edge detection of the plurality of blood-cells; (El-Zehiry et al., Figs. 1, 2, 5 & 6, Pg. 2 ¶ 0031 - Pg. 3 ¶ 0032, Pg. 3 ¶ 0039, Pg. 3 ¶ 0042 - Pg. 4 ¶ 0043) determining a bounding box for each of the plurality of blood-cells based on the contours of each of the plurality of blood-cells; (El-Zehiry et al., Figs. 1, 2, 5 & 6, Pg. 2 ¶ 0031 - Pg. 3 ¶ 0032, Pg. 3 ¶ 0039, Pg. 3 ¶ 0042 - Pg. 4 ¶ 0043) determining a count of WBCs, a count of RBCs and volumetric information of the RBCs and the WBCs based on the classification and the contours of each of the plurality of blood-cells; (El-Zehiry et al., Figs. 1, 2, 5 & 6, Pg. 2 ¶ 0030 - 0031, Pg. 3 ¶ 0038 - 0042, Pg. 5 ¶ 0055 - 0059) and outputting a report comprising the count of WBCs, the count of RBCs or the volumetric information of the RBCs and the WBCs. (El-Zehiry et al., Fig. 14, Pg. 3 ¶ 0038 - 0042, Pg. 5 ¶ 0055 - 0059) El-Zehiry et al. fail to disclose explicitly generating a first set of samples corresponding to WBCs and a second set of samples corresponding to RBCs based on the at least one WBC image and the at least one RBC image respectively using a generator model of the deep learning model, wherein the first set of samples and the second set of samples are generated based on a classification of each sample as one of real sample or fake sample using a discriminator model of the deep learning model, wherein samples corresponding to the first set of samples and the second set of samples are generated until the first set of samples is balanced with respect to the second set of samples and each of the first set of samples and the second set of samples are classified as real by the discriminator model; and wherein the plurality of images of WBCs and RBCs are determined based on the first set of samples and the second set of samples, respectively. Pertaining to analogous art, Jung et al. disclose wherein the training data is generated by: inputting at least one training image to the deep learning model, wherein the at least one training image comprises at least one WBC and/or at least one RBC; (Jung et al., Pg. 1 Abstract, Pg. 2 Subsection “Contributions”, Pg. 5 Subsection “Pre-processing of WBC images”, Pg. 5 Figs. 1 & 2, Pg. 10 Subsection “Dataset sharing” - Pg. 12 Subsection “Conclusion”, Pg. 11 Fig. 4) and determining the plurality of images of WBCs and RBCs by: generating a first set of samples corresponding to WBCs and a second set of samples corresponding to RBCs based on the at least one WBC image and the at least one RBC image respectively using a generator model of the deep learning model, (Jung et al., Pg. 1 Abstract, Pg. 2 Subsection “Contributions”, Pg. 5 Subsection “Pre-processing of WBC images”, Pg. 5 Figs. 1 & 2, Pg. 10 Subsection “Dataset sharing” - Pg. 12 Subsection “Conclusion”, Pg. 11 Fig. 4) wherein the first set of samples and the second set of samples are generated based on a classification of each sample as one of real sample or fake sample using a discriminator model of the deep learning model, (Jung et al., Pg. 1 Abstract, Pg. 2 Subsection “Contributions”, Pg. 5 Subsection “Pre-processing of WBC images”, Pg. 5 Figs. 1 & 2, Pg. 10 Subsection “Dataset sharing” - Pg. 12 Subsection “Conclusion”, Pg. 11 Fig. 4) wherein samples corresponding to the first set of samples and the second set of samples are generated until the first set of samples is balanced with respect to the second set of samples; (Jung et al., Pg. 1 Abstract, Pg. 2 Subsection “Contributions”, Pg. 5 Subsection “Pre-processing of WBC images”, Pg. 5 Figs. 1 & 2, Pg. 10 Subsection “Dataset sharing” - Pg. 12 Subsection “Conclusion”, Pg. 11 Fig. 4) and wherein the plurality of images of WBCs and RBCs are determined based on the first set of samples and the second set of samples, respectively. (Jung et al., Pg. 1 Abstract, Pg. 2 Subsection “Contributions”, Pg. 5 Subsection “Pre-processing of WBC images”, Pg. 5 Figs. 1 & 2, Pg. 10 Subsection “Dataset sharing” - Pg. 12 Subsection “Conclusion”, Pg. 11 Fig. 4) Jung et al. fail to disclose expressly wherein each of the first set of samples and the second set of samples are classified as real by the discriminator model. Pertaining to analogous art, Soni et al. disclose wherein the training data is generated by: generating a first set of samples and a second set of samples using a generator model of the deep learning model, (Soni et al., Abstract, Figs. 1, 3 - 6 & 10, Pg. 2 ¶ 0019, Pg. 3 ¶ 0026, Pg. 3 ¶ 0030, Pg. 4 ¶ 0040 - 0041, Pg. 5 ¶ 0047, Pg. 6 ¶ 0058 - Pg. 7 ¶ 0060) wherein the first set of samples and the second set of samples are generated based on a classification of each sample as one of real sample or fake sample using a discriminator model of the deep learning model, (Soni et al., Abstract, Figs. 1, 3 - 6 & 10, Pg. 2 ¶ 0019, Pg. 3 ¶ 0026 and 0030, Pg. 4 ¶ 0036 and 0038 - 0041, Pg. 5 ¶ 0043 and 0047, Pg. 6 ¶ 0058 - Pg. 7 ¶ 0060) wherein each of the first set of samples and the second set of samples are classified as real by the discriminator model; (Soni et al., Figs. 1, 3 - 6, 9 & 10, Pg. 2 ¶ 0019, Pg. 2 ¶ 0025 - Pg. 3 ¶ 0026, Pg. 4 ¶ 0036 and 0038, Pg. 4 ¶ 0040 - Pg. 5 ¶ 0043, Pg. 5 ¶ 0047, Pg. 6 ¶ 0052 and 0057 - 0058) and wherein the plurality of images are determined based on the first set of samples and the second set of samples. (Soni et al., Abstract, Figs. 1, 3 - 6, 9 & 10, Pg. 2 ¶ 0019, Pg. 2 ¶ 0025 - Pg. 3 ¶ 0026, Pg. 4 ¶ 0040 - 0041, Pg. 5 ¶ 0043 and 0047, Pg. 6 ¶ 0052, Pg. 6 ¶ 0058 - Pg. 7 ¶ 0060) Tandon et al. and El-Zehiry et al. are combinable because they are both directed towards image processing systems that process, classify and analyze blood cells. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the teachings of Tandon et al. with the teachings of El-Zehiry et al. This modification would have been prompted in order to enhance the base device of Tandon et al. with the well-known and applicable technique El-Zehiry et al. applied to a comparable device. Determining volumetric information of the RBCs and the WBCs, as taught by El-Zehiry et al., would enhance the base device of Tandon et al. by providing it with additional information that it can use when classifying the health condition of a blood sample so as to improve the overall quality and reliability of any automated diagnoses it makes. This combination could be completed according to well-known techniques in the art and would likely yield predictable results, in that volumetric information of the RBCs and the WBCs would be determined along with counts of the RBCs and the WBCs so as to provide the base device of Tandon et al. and its end-users with as much information as possible when evaluating the health condition of a blood sample. In addition, Tandon et al. in view of El-Zehiry et al. and Jung et al. are combinable because they are all directed towards image processing systems that process and classify blood cells. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combined teachings of Tandon et al. in view of El-Zehiry et al. with the teachings of Jung et al. This modification would have been prompted in order to enhance the combined base device of Tandon et al. in view of El-Zehiry et al. with the well-known technique Jung et al. applied to a comparable device. Utilizing a generator model of the deep learning model to generate training data until the first set of samples is balanced with respect to the second set of samples, as taught by Jung et al., would enhance the combined base device by improving its ability to produce a deep learning model this able to accurately and reliably classify blood cells since it would be able to easily obtain enough training images for each category of object to be classified to ensure that the deep learning model is properly trained. Furthermore, this modification would have been prompted by the teachings and suggestions of El-Zehiry et al. that the training set used to train their machine learning model had 100 images per category, see at least page 5 paragraph 0053 of El-Zehiry et al. This combination could be completed according to well-known techniques in the art and would likely yield predictable results, in that a generator model of the deep learning model would be utilized to generate training data until the first set of samples is balanced with respect to the second set of samples so as to ensure that sufficient examples of each category of object to be classified by the deep learning model is readily and easily available to ensure that the deep learning model is properly trained to provide accurate and reliable classification of blood cell images. Additionally, Tandon et al. in view of El-Zehiry et al. in view of Jung et al. and Soni et al. are combinable because they are all directed towards image processing systems that utilize machine learning models to evaluate medical images. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combined teachings of Tandon et al. in view of El-Zehiry et al. in view of Jung et al. with the teachings of Soni et al. This modification would have been prompted in order to enhance the combined base device of Tandon et al. in view of El-Zehiry et al. in view of Jung et al. with the well-known technique Soni et al. applied to a similar device. Utilizing first and second sets of samples classified as real by a discriminator model as training data, as taught by Soni et al., would enhance the combined base device by improving its ability to accurately and reliably classify blood cells since only the most realistic and convincing looking synthetically generated images would be utilized when training the deep learning model. Furthermore, this modification would have been prompted by the teachings and suggestions of Jung et al. that the synthetic blood cell images they generated using their trained Deep Convolutional Generative Adversarial Network (DCGAN) were published so they could be used for education and research purposes and that the synthetic images they generated were verified using four techniques including having a medical laboratory specialist classify the synthetic blood cell images without knowing they were synthetically generated, see at least page 2 subsection “Contributions” and page 10 subsection “Dataset sharing” - page 11 subsection “Generated image quality” of Jung et al. This combination could be completed according to well-known techniques in the art and would likely yield predictable results, in that only first and second sets of samples classified as real by a discriminator model would be utilized as the training data so as to ensure that only the highest quality synthetically generated images are used to train the deep learning model of the combined base device. Therefore, it would have been obvious to combine Tandon et al. with El-Zehiry et al., Jung et al. and Soni et al. to obtain the invention as specified in claims 1, 8 and 15.
- With regards to claims 3, 10 and 17, Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. disclose the method, system and non-transitory computer-readable medium of claims 1, 8 and 15, respectively, wherein the edges of each of the plurality of blood cells are determined by determining a bimodal image of the blood smear image upon the preprocessing, (Tandon et al., Figs. 10 - 12, Pg. 6 ¶ 0146 - 0147, Pg. 8 ¶ 0166 - 0169, Pg. 14 ¶ 0262 - 0268, Pg. 15 ¶ 0274 - 0276, Pg. 16 ¶ 0284 - 0290) wherein the edges of each of the plurality of blood cells are determined based on the bimodal image by using an edge detection technique. (Tandon et al., Figs. 10 - 12, Pg. 6 ¶ 0146 - 0147, Pg. 8 ¶ 0166 - 0169, Pg. 14 ¶ 0262 - 0268, Pg. 15 ¶ 0274 - 0276, Pg. 16 ¶ 0284 - 0290)
- With regards to claims 4, 11 and 18, Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. disclose the method, system and non-transitory computer-readable medium of claims 1, 8 and 15, respectively, further comprises: determining, by the processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) bounding boxes for each of the at least one WBC and/or the at least one RBC based on detection of contours of the at least one WBC and the at least one RBC in the at least one training image; (Tandon et al., Figs. 23 - 25, Pg. 6 ¶ 0146 - 0147, Pg. 14 ¶ 0262 - 0263, Pg. 16 ¶ 0285 - 0290, Pg. 18 ¶ 0330 - 0332) and cropping, by the processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) each of the bounding boxes to determine at least one WBC image and at least one RBC image. (Tandon et al., Figs. 23 - 25, Pg. 6 ¶ 0146 - 0147, Pg. 12 ¶ 0212 - 0219, Pg. 13 ¶ 0221, Pg. 14 ¶ 0253 - 0259, Pg. 16 ¶ 0285 - 0296, Pg. 18 ¶ 0330 - 0334) In addition, analogous art Jung et al. disclose determining bounding boxes for each of the at least one WBC and/or the at least one RBC based on the at least one WBC and the at least one RBC in the at least one training image; (Jung et al., Pg. 1 Abstract, Pg. 2 Subsection “Contributions”, Pg. 5 Subsection “Pre-processing of WBC images”, Pg. 5 Figs. 1 & 2, Pg. 10 Subsection “Dataset sharing” - Pg. 12 Subsection “Conclusion”, Pg. 11 Fig. 4) and cropping each of the bounding boxes to determine at least one WBC image and at least one RBC image. (Jung et al., Pg. 1 Abstract, Pg. 2 Subsection “Contributions”, Pg. 5 Subsection “Pre-processing of WBC images”, Pg. 5 Figs. 1 & 2, Pg. 10 Subsection “Dataset sharing” - Pg. 12 Subsection “Conclusion”, Pg. 11 Fig. 4)
- With regards to claims 7 and 14, Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. disclose the method and system of claims 1 and 8, respectively, comprising: classifying, by the processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) each of the plurality of blood-cells classified as the WBC as one of a plurality of WBC classes using the deep learning model, wherein each of the plurality of WBC classes correspond to a type of WBC from a plurality of WBC types. (Tandon et al., Fig. 25, Pg. 7 ¶ 0150 - 0152, Pg. 12 ¶ 0218 - 0219, Pg. 13 ¶ 0221 and 0231 - 0236, Pg. 16 ¶ 0296 - Pg. 17 ¶ 0298, Pg. 17 ¶ 0306 - 0313, Pg. 21 ¶ 0363 - 0368)
Claims 2, 9 and 16 are rejected under 35 U.S.C. 103 as being unpatentable over Tandon et al. U.S. Publication No. 2018/0211380 A1 in view of El-Zehiry et al. U.S. Publication No. 2017/0132450 A1 in view of Jung et al., "WBC image classification and generative models based on convolutional neural network", BMC Medical Imaging, Vol. 22, No. 94, May 2022, pages 1 - 16 in view of Soni et al. U.S. Publication No. 2020/0311913 A1 as applied to claims 1, 8 and 15 above, and further in view of Murphy et al. U.S. Publication No. 2018/0328848 A1.
- With regards to claims 2, 9 and 16, Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. disclose the method, system and non-transitory computer-readable medium of claims 1, 8 and 15, respectively, wherein the preprocessing comprises: enhancing contrast, by the processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) of the blood smear image; (Tandon et al., Figs. 10 - 12, Pg. 6 ¶ 0146 - 0147, Pg. 8 ¶ 0166 - 0169, Pg. 14 ¶ 0262 - 0268, Pg. 15 ¶ 0274 - 0276, Pg. 16 ¶ 0284 - 0290) and removing noise, by the processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) in the blood smear image. (Tandon et al., Pg. 14 ¶ 0262 - 0268) Tandon et al. fail to disclose explicitly using a histogram equalization technique; and removing noise in the contrast enhanced blood smear image using a gaussian filter. Pertaining to analogous art, Murphy et al. disclose wherein the preprocessing comprises: removing noise in the contrast enhanced blood smear image using a gaussian filter. (Murphy et al., Figs. 4, 5 & 7A, Pg. 7 ¶ 0083 - 0086 and 0088 - 0089) Murphy et al. fail to disclose explicitly using a histogram equalization technique. However, the Examiner takes official notice of the fact that utilizing a histogram equalization technique when preprocessing digital images is notoriously well-known in the art. Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. and Murphy et al. are combinable because they are all directed towards image processing systems that evaluate medical images and, similar to Tandon et al., El-Zehiry et al. and Jung et al., Murphy et al. is also directed towards processing, analyzing and classifying blood cells. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combined teachings of Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. with the teachings of Murphy et al. This modification would have been prompted in order to enhance the combined base device of Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. with the well-known and applicable technique Murphy et al. applied to a similar device. Removing noise in the contrast enhanced blood smear image using a gaussian filter, as taught by Murphy et al., would enhance the combined base device by improving its ability to accurately and reliably classify and analyze images of blood cells since as much erroneous image data as possible would be removed from the images and thus prevented from affecting classification and analysis results of the combined base device. This combination could be completed according to well-known techniques in the art and would likely yield predictable results, in that a gaussian filter would be utilized to remove noise from the contrast enhanced blood smear image so as to improve the ability of the combined base device to accurately and reliably classify and analyze images of blood cells. In addition, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combined teachings of Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. in view of Murphy et al. to include a histogram equalization technique during image preprocessing. This modification would have been prompted in order to enhance the ability of the combined base device of Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. in view of Murphy et al. with the notoriously well-known technique of utilizing histogram equalization to enhance the quality of images and improve discrimination between imaged objects. Utilizing a histogram equalization technique to enhance the contrast of the blood smear image would enhance the combined base device improving the contrast of the blood smear images such that imaged blood cells stand out more from the background thereby improving the ability of the combined base device to accurately and reliably detect, segment and analyze individual blood cells in the blood smear images. This combination could be completed according to well-known techniques in the art and would likely yield predictable results, in that a histogram equalization technique would be utilized to enhance the quality of the blood smear images processed by the combined base device. Therefore, it would have been obvious to combine Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. with Murphy et al. and the notoriously well-known technique of utilizing a histogram equalization technique to preprocess digital images to obtain the invention as specified in claims 2, 9 and 16. The Examiner asserts that the common knowledge or well-known in the art statement is taken to be admitted prior art because applicant failed to traverse the examiner’s assertion of official notice.
Claims 5, 6, 12, 13, 19 and 20 are rejected under 35 U.S.C. 103 as being unpatentable over Tandon et al. U.S. Publication No. 2018/0211380 A1 in view of El-Zehiry et al. U.S. Publication No. 2017/0132450 A1 in view of Jung et al., "WBC image classification and generative models based on convolutional neural network", BMC Medical Imaging, Vol. 22, No. 94, May 2022, pages 1 - 16 in view of Soni et al. U.S. Publication No. 2020/0311913 A1 as applied to claims 1, 8 and 15 above, and further in view of Park et al. U.S. Publication No. 2023/0030787 A1.
- With regards to claims 5, 12 and 19, Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. disclose the method, system and non-transitory computer-readable medium of claims 1, 8 and 15, respectively, comprises: displaying, by the processor, (Tandon et al., Fig. 4A, Pg. 3 ¶ 0038, Pg. 10 ¶ 0194 - 0197, Pg. 22 ¶ 0381 - Pg. 23 ¶ 0388) an analysis received from the system on a display screen. (Tandon et al., Pg. 7 ¶ 0150 - 0152, Pg. 8 ¶ 0162 - 0166, Pg. 11 ¶ 0205 - Pg. 12 ¶ 0211, Pg. 16 ¶ 0292 - 0293, Pg. 21 ¶ 0362 - 0365) Tandon et al. fail to disclose explicitly inputting the report as a query to a generative artificial intelligence-based query system; and an analysis received from the generative artificial intelligence-based query system. Pertaining to analogous art, Park et al. disclose inputting the report as a query to a generative artificial intelligence-based query system; (Park et al., Abstract, Figs. 2 - 5, Pg. 2 ¶ 0027 - 0028, Pg. 3 ¶ 0036 - 0044, Pg. 4 ¶ 0053 - 0060, Pg. 4 ¶ 0063 - Pg. 5 ¶ 0071) and displaying an analysis received from the generative artificial intelligence-based query system on a display screen. (Park et al., Abstract, Figs. 2 - 5, Pg. 2 ¶ 0027 - 0028, Pg. 3 ¶ 0036 - 0044, Pg. 4 ¶ 0053 - 0060, Pg. 4 ¶ 0063 - Pg. 5 ¶ 0071) Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. and Park et al. are combinable because they are all directed towards systems and methods that utilize machine learning models to evaluate the health status of an individual and, similar to Tandon et al., El-Zehiry et al. and Jung et al., Park et al. is also directed towards utilizing WBC counts to evaluate the health status of an individual. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the combined teachings of Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. with the teachings of Park et al. This modification would have been prompted in order to enhance the combined base device of Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. with the well-known technique Park et al. applied to a comparable device. Inputting the report as a query to a generative artificial intelligence-based query system, as taught by Park et al., would enhance the combined base device by allowing for end-users to utilize the WBC and RBC metrics it generated for various other purposes and in an increased number of applications, such as by inputting them into additional machine learning models, in order to improve the overall usefulness and appeal of the combined base device to potential end-users. This combination could be completed according to well-known techniques in the art and would likely yield predictable results, in that the report generated by the combined base device would be input as a query to a generative artificial intelligence-based query system so as to allow for end-users to utilize the WBC and RBC metrics it generated for various other purposes and in an increased number of applications and thereby improve the overall usefulness and appeal of the combined base device to potential end-users. Therefore, it would have been obvious to combine Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. with Park et al. to obtain the invention as specified in claims 5, 12 and 19.
- With regards to claims 6, 13 and 20, Tandon et al. in view of El-Zehiry et al. in view of Jung et al. in view of Soni et al. in view of Park et al. disclose the method, system and non-transitory computer-readable medium of claims 5, 12 and 19, respectively, wherein the analysis is based on the count of WBCs, the count of RBCs or the volumetric information of the RBCs and WBCs, (Tandon et al., Abstract, Pg. 21 ¶ 0363 - Pg. 22 ¶ 0379) wherein the analysis comprises one or more health conditions determined based on a comparison of the count of WBCs, the count of RBCs or the volumetric information of the RBCs and the WBCs with a corresponding predefined threshold count of WBCs, a predefined threshold count of RBCs or a predefined volumetric threshold of RBCs and WBCs. (Tandon et al., Abstract, Pg. 21 ¶ 0363 - Pg. 22 ¶ 0379)
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Lorsakul et al. U.S. Publication No. 2023/0230242 A1; which is directed towards a system and method for analyzing digital pathology images, wherein a new image is generated by a Generative Adversarial Network (GAN), the new image is judged as being either authentic or fake and, when the new image is judged as being authentic, the new image is output, see at least the abstract and figure 7 of Lorsakul et al.
Madani et al. U.S. Publication No. 2019/0197358 A1; which is directed towards a medical image processing system and method, wherein a generator of a Generative Adversarial Network (GAN) is trained to generate synthetic medical images that approximate real medical images to a level that a discriminator of the GAN is fooled by the generated medical images so as to permit the generator to be used as an additional source of unlabeled medical images that may be used to expand a training medical image data set, see at least the abstract, page 2 paragraphs 0022 and 0027, page 3 paragraphs 0031 and 0033, and page 4 paragraph 0037 of Madani et al.
Park et al. U.S. Publication No. 2024/0087265 A1; which is directed towards an image processing system and method for creating or modifying images, wherein a generator of a Generative Adversarial Network (GAN) iteratively generates synthetic images until a synthetic image is generated that fools a discriminator of the GAN, at which point the corresponding synthetic image is output, see at least page 12 paragraph 0094 and page 13 paragraph 0106 of Park et al.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ERIC RUSH whose telephone number is (571) 270-3017. The examiner can normally be reached 9am - 5pm 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 http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Andrew Bee can be reached at (571) 270 - 5183. 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.
/ERIC RUSH/Primary Examiner, Art Unit 2677
