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 .
DETAILED ACTION
The United States Patent & Trademark Office appreciates the application that is submitted by the inventor/assignee. The United States Patent & Trademark Office reviewed the following application and has made the following comments below.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 06/24/2024 is considered and attached.
Priority
This application claims provisional benefit of 63484674 filed on 02/13/2023.
Claim Status
Claim 1 is rejected under 35 USC § 102 in view of Li.
Claims 2-5, 7-8, 11-16 and 19-20 are rejected under 35 USC § 103:
Claims 2, 8 and 13-14 are rejected over Li in view of Cumming.
Claims 3, 5, 7, 15, 17 and 19-20 are rejected over Li in view of Cumming in view of Suwannaphong in view of Pollak.
Claims 4 and 16 are rejected over Li in view of Cumming in view of Suwannaphong in view of Pollak in view of Nguyen.
Claim 11 is rejected over Li in view of Cumming in view of Huang.
Claim 12 is rejected over Li in view of Cumming in view of Cheng.
Claims 6, 9-10 and 18 are objected.
Claim Interpretation
The following is a quotation of 35 U.S.C. 112(f):
(f) Element in Claim for a Combination. – An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The following is a quotation of pre-AIA 35 U.S.C. 112, sixth paragraph:
An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.
The claims in this application are given their broadest reasonable interpretation using the plain meaning of the claim language in light of the specification as it would be understood by one of ordinary skill in the art. The broadest reasonable interpretation of a claim element (also commonly referred to as a claim limitation) is limited by the description in the specification when 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is invoked.
As explained in MPEP § 2181, subsection I, claim limitations that meet the following three-prong test will be interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph:
(A) the claim limitation uses the term “means” or “step” or a term used as a substitute for “means” that is a generic placeholder (also called a nonce term or a non-structural term having no specific structural meaning) for performing the claimed function;
(B) the term “means” or “step” or the generic placeholder is modified by functional language, typically, but not always linked by the transition word “for” (e.g., “means for”) or another linking word or phrase, such as “configured to” or “so that”; and
(C) the term “means” or “step” or the generic placeholder is not modified by sufficient structure, material, or acts for performing the claimed function.
Use of the word “means” (or “step”) in a claim with functional language creates a rebuttable presumption that the claim limitation is to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites sufficient structure, material, or acts to entirely perform the recited function.
Absence of the word “means” (or “step”) in a claim creates a rebuttable presumption that the claim limitation is not to be treated in accordance with 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph. The presumption that the claim limitation is not interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, is rebutted when the claim limitation recites function without reciting sufficient structure, material or acts to entirely perform the recited function.
This application includes one or more claim limitations that do not use the word “means,” but are nonetheless being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, because the claim limitation(s) uses a generic placeholder that is coupled with functional language without reciting sufficient structure to perform the recited function and the generic placeholder is not preceded by a structural modifier. Such claim limitation(s) is/are:
“a chamber configured to …” in claim 1. The corresponding structure is “a cell counter slide chamber” in Specification, [0043].
“a light source configured to …” in claim 1. The corresponding structure is “a light source 210” in Specification, [0026].
Because this/these claim limitation(s) is/are being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, it/they is/are being interpreted to cover the corresponding structure described in the specification as performing the claimed function, and equivalents thereof.
If applicant does not intend to have this/these limitation(s) interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph, applicant may: (1) amend the claim limitation(s) to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph (e.g., by reciting sufficient structure to perform the claimed function); or (2) present a sufficient showing that the claim limitation(s) recite(s) sufficient structure to perform the claimed function so as to avoid it/them being interpreted under 35 U.S.C. 112(f) or pre-AIA 35 U.S.C. 112, sixth paragraph.
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.
CLAIM 1
Claims 1 are rejected under 35 U.S.C. 102 (a)(1) as being anticipated by Li et al. (Li, Yaning, et al. "A low‐cost, automated parasite diagnostic system via a portable, robotic microscope and deep learning." Journal of biophotonics 12.9, published 2019, hereinafter Li)
In regards to Claim 1, Li teaches a device (Li, page 2, right col, last paragraph: “we propose a solution that combines a low-cost, automated microscope system and convolutional neural networks to image samples simply prepared using the standard McMaster floatation method, accurately extracting egg counts as well as providing diagnosis based on extracted egg morphologies”; see modified figure 1 below) comprising:
a chamber configured to receive a sample specimen (Li, Abstract: “Fecal samples prepared using the McMaster flotation method were imaged, with the imaging region comprising the entire McMaster chamber”);
a light source configured to illuminate a field of view of the sample specimen in the chamber (Li, page 3, right col, second paragraph: “The sample is illuminated from below by a diffuse white led, approximating a Köhler illumination setup”; see modified figure 1 below);
PNG
media_image1.png
770
1158
media_image1.png
Greyscale
a microscope objective to magnify the field of view of the sample specimen; and
a camera configured to image the field of view of the specimen sample (Li, page 3, right col, section 2.2: “compared to the prior report, the microscope has been adapted for the purpose of egg imaging, as shown in Figure 1. First, the optical system was altered to be upright instead of inverted to accommodate the floatation solution prepared samples”; see modified figure 1 below),
to produce an on-site dataset of images (Li, Page 4, left col, first paragraph: “The sample is automatically scanned, and 18 × 24 images are acquired, imaging the entire McMaster grid. In the deep learning analysis described below, individual images are analyzed”), and to provide the on-site dataset of images to a trained machine-learning model (Li, page 4, section 2.2.2: “The CNN is based on a U-Net structure”; Figure 3: “Once the network has been trained and validated, new data inputted into the network will yield preliminary segmentation results”) for analysis of at least one species of parasites. (Li, page 6, right col, last paragraph: “eggs were classified as Coccidia or nematodes based on their size”).
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.
Claim(s) 2, 8 and 13-14 is/are rejected under 35 U.S.C. 103 as being unpatentable over Li in view of Cumming et al. (US-20210248419-A1, hereinafter Cumming).
CLAIM 2
Regarding claim 2, Li teaches the device of Claim 1. In addition, Li teaches a processor (Li, see laptop in figure 1 A; page 4-5, section 2.2.1 and 2.2.2. The Examiner notes the training and applying a CNN implies a computing device with processor, memory and storage); and
a memory that includes instructions executable by the processor for causing the processor to perform operations (Li, see laptop in figure 1 A; page 4-5, section 2.2.1 and 2.2.2. The Examiner notes the training and applying a CNN implies a computing device with processor, memory and storage) comprising:
receiving a dataset of labeled images, each labeled image in the dataset of labeled images categorized as an image with zero oocytes, an image with oocytes of a parasite species in isolation, or (***The Examiner notes since a listing with “or” is disjunctive, any one of the elements found in the prior art is sufficient to reject the claim. While citations have been provided for completeness and rapid prosecution, only one element is required.) an image with oocytes of parasite species in combination (Li, page 4, section 2.2.2: “The whole dataset for CNN construction consists of 951 images (643 labeled images with eggs and 308 images without eggs)”);
Li does not explicitly disclose performing at least one cropping or at least one rotation to each image in the dataset of labeled images to produced an expanded training dataset;
Cumming is in the same field of art of identification and quantification parasite eggs in fecal samples using machine learning model. Further, Cumming teaches performing at least one cropping or (***The Examiner notes since a listing with “or” is disjunctive, any one of the elements found in the prior art is sufficient to reject the claim. While citations have been provided for completeness and rapid prosecution, only one element is required.) at least one rotation to each image in the dataset of labeled images to produced an expanded training dataset (Cumming, ¶ [0070]: “cropping one or more of the plurality of training images by human and/or computer means so as to produce one or more cropped images”);
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Li by incorporating method to generate additional training dataset by cropping technique that is taught by Cumming, to make a machine learning based system that can generate additional training dataset; thus, one of ordinary skilled in the art would be motivated to combine the references since among its several aspects, the present invention recognizes there is a need to improve performance of machine learning model by increasing training dataset volume (Cumming, ¶ [0239]: “It is contemplated that performance of the algorithm may be improved by training with a larger number of training images and/or a more diverse range of training images”).
The combination of Li and Cumming then teaches training, using the expanded training dataset, a machine-learning model into a trained machine-learning model that determines speciation of at least one parasite species (Li, page 6, right col, last paragraph: “eggs were classified as Coccidia or nematodes based on their size) (Cumming, ¶ [0054]: “The cropped images are used as input for a machine learning model for training to identify the ovum species within each cropped frame”), a count of oocytes of the at least one parasite species (Li, page 4, section 2.2.1: “Once images have been acquired, they must be automatically analyzed to identify candidate eggs for diagnosis and counting”; page 7, section 3.2: “After the combined neural network image segmentation and morphological discrimination, final counts of numbers of eggs per sample are obtained”), and an infection status of the at least one parasite species. (Cumming, ¶ [0065]: “The present invention is predicated, in one aspect, at least in part on the finding that machine learning techniques may be used to identify the presence or absence of target biological material”)
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
CLAIM 8
Regarding claim 8, Li teaches a method (Li, page 2, right col, last paragraph: “we propose a solution that combines a low-cost, automated microscope system and convolutional neural networks to image samples simply prepared using the standard McMaster floatation method, accurately extracting egg counts as well as providing diagnosis based on extracted egg morphologies”) comprising: receiving a dataset of labeled images, each labeled image in the dataset of labeled images categorized as an image with zero oocytes, an image with oocytes of a parasite species in isolation, or (***The Examiner notes since a listing with “or” is disjunctive, any one of the elements found in the prior art is sufficient to reject the claim. While citations have been provided for completeness and rapid prosecution, only one element is required.) an image with oocytes of parasite species in combination; (Li, page 4, section 2.2.2: “The whole dataset for CNN construction consists of 951 images (643 labeled images with eggs and 308 images without eggs)”)
Li does not explicitly disclose performing at least one cropping or at least one rotation to each image in the dataset of labeled images to produce an expanded training dataset;
Cumming is in the same field of art of identification and quantification parasite eggs in fecal samples using machine learning model. Further, Cumming teaches performing at least one cropping or (***The Examiner notes since a listing with “or” is disjunctive, any one of the elements found in the prior art is sufficient to reject the claim. While citations have been provided for completeness and rapid prosecution, only one element is required.) at least one rotation to each image in the dataset of labeled images to produce an expanded training dataset (Cumming, ¶ [0070]: “cropping one or more of the plurality of training images by human and/or computer means so as to produce one or more cropped images”);
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Li by incorporating method to generate additional training dataset by cropping technique that is taught by Cumming, to make a machine learning based system that can generate additional training dataset; thus, one of ordinary skilled in the art would be motivated to combine the references since among its several aspects, the present invention recognizes there is a need to improve performance of machine learning model by increasing training dataset volume (Cumming, ¶ [0239]: “It is contemplated that performance of the algorithm may be improved by training with a larger number of training images and/or a more diverse range of training images”).
The combination of Li and Cumming then teaches training, using the expanded training dataset, a machine-learning model into a trained machine-learning model that determines speciation of at least one parasite species (Li, page 6, right col, last paragraph: “eggs were classified as Coccidia or nematodes based on their size) (Cumming, ¶ [0054]: “The cropped images are used as input for a machine learning model for training to identify the ovum species within each cropped frame”), a count of oocytes of the at least one parasite species (Li, page 4, section 2.2.1: “Once images have been acquired, they must be automatically analyzed to identify candidate eggs for diagnosis and counting”; page 7, section 3.2: “After the combined neural network image segmentation and morphological discrimination, final counts of numbers of eggs per sample are obtained”), and an infection status of the at least one parasite species. (Cumming, ¶ [0065]: “The present invention is predicated, in one aspect, at least in part on the finding that machine learning techniques may be used to identify the presence or absence of target biological material”)
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
CLAIM 13
Regarding Claim 13, the combination of Li and Cumming teaches the method of Claim 8. In addition, the combination of Li and Cumming teaches performing at least one cropping comprises cropping in a sliding window fashion (Cumming, ¶ [0117-0119]: “an ovum frame is determined by a method such as a sliding window-based method”) by repeatedly changing, for each image, a starting location of a crop by incrementing an offset in one or more directions on the image. (Cumming, ¶ [0117-0119]: “A window of fixed dimension (width and height) is used to slide through the entire input image … At each position of the window on the input image, an image frame, corresponding to the current window, is cropped out …”, see FIG. 4, the window is slid from left to right)
CLAIM 14
Regarding Claim 14, the combination of Li and Cumming teaches the method of Claim 13. In addition, the combination of Li and Cumming teaches performing at least one cropping in the sliding window fashion comprises cropping in fixed window dimensions at regular increments from a top-left of an input image to a bottom-right of the input image. (Cumming, ¶ [0117-0119]: “A window of fixed dimension (width and height) is used to slide through the entire input image … At each position of the window on the input image, an image frame, corresponding to the current window, is cropped out …”, see FIG. 4, first window starts at the top-left, and is slid to the right-bottom of the image)
Claim(s) 3, 5, 7, 15, 17 and 19-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Li in view of Cumming in view Suwannaphong et al. (Suwannaphong, Thanaphon, et al. "Parasitic Egg Detection and Classification in Low-cost Microscopic Images using Transfer Learning." arXiv preprint arXiv:2107.00968, published 2021, hereinafter Suwannaphong), and further in view of Pollak et al. (US-20230159982-A1, cited in IDS, filed 01/23/2023, hereinafter Pollak).
CLAIM 3
Regarding Claim 3, Li teaches the device of Claim 1. In addition, Li teaches a processor (Li, see laptop in figure 1 A; page 4-5, section 2.2.1 and 2.2.2. The Examiner notes the training and applying a CNN implies a computing device with processor, memory and storage); and
a memory that includes instructions executable by the processor for causing the processor to perform operations (Li, see laptop in figure 1 A; page 4-5, section 2.2.1 and 2.2.2. The Examiner notes the training and applying a CNN implies a computing device with processor, memory and storage) comprising:
receiving the on-site dataset of images; (Li, Page 4, left col, first paragraph: “The sample is automatically scanned, and 18 × 24 images are acquired, imaging the entire McMaster grid. In the deep learning analysis described below, individual images are analyzed”)
accessing a trained machine-learning model; (Li, page 4, section 2.2.2: “The CNN is based on a U-Net structure”; Figure 3: “Once the network has been trained and validated, new data inputted into the network will yield preliminary segmentation results”)
Li does not explicitly disclose performing at least one cropping on the on-site dataset to produce an expanded on-site dataset;
Cumming is in the same field of art of identification and quantification parasite eggs in fecal samples using machine learning model. Further, Cumming teaches performing at least one cropping on the on-site dataset to produce an expanded on-site dataset (Cumming, ¶ [0070]: “cropping one or more of the plurality of training images by human and/or computer means so as to produce one or more cropped images”);
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Li by incorporating method to generate additional training dataset by cropping technique that is taught by Cumming, to make a machine learning based system that can generate additional training dataset; thus, one of ordinary skilled in the art would be motivated to combine the references since among its several aspects, the present invention recognizes there is a need to improve performance of machine learning model by increasing training dataset volume (Cumming, ¶ [0239]: “It is contemplated that performance of the algorithm may be improved by training with a larger number of training images and/or a more diverse range of training images”).
The combination of Li and Cumming then teaches applying the trained machine-learning model to the expanded on-site dataset to produce an evaluated on-site dataset; (Cumming ¶ [0094]: “Potential ova are identified from the parasite ovum images, and a deep convolutional neural network applied to extract features (and preferably identification-relevant features), and the features then used to determine the ovum species by reference to a reference database. The detection outcomes are then stored in a database where ovum species tallying may be queried.”)
The combination of Li and Cumming does not explicitly disclose reconstructing the evaluated on-site dataset.
Suwannaphong is in the same field of art of parasite egg detecting using machine learning model. Further, Suwannaphong teaches reconstructing the evaluated on-site dataset. (Suwannaphong, page 2, section 2.2.1: “Each microscopic image is divided into patches, allowing the model to characterise the whole image by analysing local areas … merging probability of all patches together to reconstruct the probability map corresponding the input microscopic image…” Suwannaphong teaches dividing an original input image to patches, analyzing each patch for parasite eggs, then merge all patches together to reconstruct the a probability map corresponding to the original image)
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Li and Cumming by incorporating the patch overlapping technique that is taught by Suwannaphong, to make a machine learning based system to detect and count parasite eggs, that utilize the patch overlapping technique for detection; thus, one of ordinary skilled in the art would be motivated to combine the references since among its several aspects, the present invention recognizes there is a need to increase true positive rates in the task detecting parasite eggs (Suwannaphong, page 4, section 3.1: “In the detection process, the probabilities of overlapping patches are merged into the probability map of the testing microscopic image. This results in an increase in the probability of there being a parasitic egg in the egg area. As a result, the true positive rates increase when analysing the whole image”).
The combination of Li, Cumming and Suwannaphong then teaches reconstructing the evaluated on-site dataset to produce an output, the output comprising an on-site speciation of at least one parasite species (Li, page 6, right col, last paragraph: “eggs were classified as Coccidia or nematodes based on their size) (Cumming, ¶ [0054]: “The cropped images are used as input for a machine learning model for training to identify the ovum species within each cropped frame”), a count of oocytes of the at least one parasite species (Li, page 4, section 2.2.1: “Once images have been acquired, they must be automatically analyzed to identify candidate eggs for diagnosis and counting”; page 7, section 3.2: “After the combined neural network image segmentation and morphological discrimination, final counts of numbers of eggs per sample are obtained”), and an infection status of the at least one parasite species; (Cumming, ¶ [0065]: “The present invention is predicated, in one aspect, at least in part on the finding that machine learning techniques may be used to identify the presence or absence of target biological material”)
The combination of Li, Cumming and Suwannaphong does not explicitly disclose sharing the output with a user.
Pollak is in the same field of art of detecting parasitic infection in biological sample. Further, Pollak teaches sharing the output with a user. (Pollak, ¶ [0159-0160]: “The processing unit is also configured to communicate with an output unit. Thus, based on the output value or values of the structural features determined by the processing unit, an output signal or output signals indicative of presence or absence of at least one pathogen and in some embodiments, the type of the pathogen, in the acquired image is provided by an output unit … including a graph, graphic or text displayed on a monitor of a control unit, a printout, as a voice message, or a user's smartphone display”)
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Li, Cumming and Suwannaphong by incorporating outputting unit that is taught by Pollak, to make a system to detect parasitic infection and sharing the result with user; thus, one of ordinary skilled in the art would be motivated to combine the references since among its several aspects, the present invention recognizes there is a need to improve user experience in such system by sharing result with user (Pollak, ¶ [0159-0160]).
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
CLAIM 5
Regarding Claim 5, the combination of Li, Cumming, Suwannaphong and Pollak teaches the device of Claim 3. In addition, the combination of Li, Cumming, Suwannaphong and Pollak teaches the infection status comprises a binary status of an infected status or a non-infected status. (Cumming, ¶ [0065]: “The present invention is predicated, in one aspect, at least in part on the finding that machine learning techniques may be used to identify the presence or absence of target biological material”)
CLAIM 7
Regarding Claim 7, the combination of Li, Cumming, Suwannaphong and Pollak teaches the device of Claim 3. In addition, the combination of Li, Cumming, Suwannaphong and Pollak teaches the operation of reconstructing the evaluated on-site dataset comprises remerging portions of cropped oocytes (Suwannaphong, page 2, section 2.2.1: “Each microscopic image is divided into patches, allowing the model to characterise the whole image by analysing local areas … merging probability of all patches together to reconstruct the probability map corresponding the input microscopic image…” Suwannaphong teaches dividing an original input image to patches, analyzing each patch for parasite eggs, then merge all patches together to reconstruct a probability map corresponding to the original image)
and correcting counts of oocytes for at least one parasite species. (Li, page 4, section 2.2.1: “Once images have been acquired, they must be automatically analyzed to identify candidate eggs for diagnosis and counting”; page 7, section 3.2: “After the combined neural network image segmentation and morphological discrimination, final counts of numbers of eggs per sample are obtained”)
CLAIM 15
Regarding claim 15, Li teaches a non-transitory computer-readable medium comprising instructions that are executable by a processor for causing the processor to perform operations (Li, see laptop in figure 1 A; page 4-5, section 2.2.1 and 2.2.2. The Examiner notes the training and applying a CNN implies a computing device with processor, memory and storage) comprising:
receiving an on-site dataset; (Li, Page 4, left col, first paragraph: “The sample is automatically scanned, and 18 × 24 images are acquired, imaging the entire McMaster grid. In the deep learning analysis described below, individual images are analyzed”)
accessing a trained machine-learning model; (Li, page 4, section 2.2.2: “The CNN is based on a U-Net structure”; Figure 3: “Once the network has been trained and validated, new data inputted into the network will yield preliminary segmentation results”)
Li does not explicitly disclose performing at least one cropping on the on-site dataset to produce an expanded on-site dataset;
Cumming is in the same field of art of identification and quantification parasite eggs in fecal samples using machine learning model. Further, Cumming teaches performing at least one cropping on the on-site dataset to produce an expanded on-site dataset (Cumming, ¶ [0070]: “cropping one or more of the plurality of training images by human and/or computer means so as to produce one or more cropped images”);
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Li by incorporating method to generate additional training dataset by cropping technique that is taught by Cumming, to make a machine learning based system that can generate additional training dataset; thus, one of ordinary skilled in the art would be motivated to combine the references since among its several aspects, the present invention recognizes there is a need to improve performance of machine learning model by increasing training dataset volume (Cumming, ¶ [0239]: “It is contemplated that performance of the algorithm may be improved by training with a larger number of training images and/or a more diverse range of training images”).
The combination of Li and Cumming then teaches applying the trained machine-learning model to the expanded on-site dataset to produce an evaluated on-site dataset; (Cumming ¶ [0094]: “Potential ova are identified from the parasite ovum images, and a deep convolutional neural network applied to extract features (and preferably identification-relevant features), and the features then used to determine the ovum species by reference to a reference database. The detection outcomes are then stored in a database where ovum species tallying may be queried.”)
The combination of Li and Cumming does not explicitly disclose reconstructing the evaluated on-site dataset.
Suwannaphong is in the same field of art of parasite egg detecting using machine learning model. Further, Suwannaphong teaches reconstructing the evaluated on-site dataset. (Suwannaphong, page 2, section 2.2.1: “Each microscopic image is divided into patches, allowing the model to characterise the whole image by analysing local areas … merging probability of all patches together to reconstruct the probability map corresponding the input microscopic image…” Suwannaphong teaches dividing an original input image to patches, analyzing each patch for parasite eggs, then merge all patches together to reconstruct the a probability map corresponding to the original image)
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Li and Cumming by incorporating the patch overlapping technique that is taught by Suwannaphong, to make a machine learning based system to detect and count parasite eggs, that utilize the patch overlapping technique for detection; thus, one of ordinary skilled in the art would be motivated to combine the references since among its several aspects, the present invention recognizes there is a need to increase true positive rates in the task detecting parasite eggs (Suwannaphong, page 4, section 3.1: “In the detection process, the probabilities of overlapping patches are merged into the probability map of the testing microscopic image. This results in an increase in the probability of there being a parasitic egg in the egg area. As a result, the true positive rates increase when analysing the whole image”).
The combination of Li, Cumming and Suwannaphong then teaches reconstructing the evaluated on-site dataset to produce an output, the output comprising an on-site speciation of at least one parasite species (Li, page 6, right col, last paragraph: “eggs were classified as Coccidia or nematodes based on their size) (Cumming, ¶ [0054]: “The cropped images are used as input for a machine learning model for training to identify the ovum species within each cropped frame”), a count of oocytes of the at least one parasite species (Li, page 4, section 2.2.1: “Once images have been acquired, they must be automatically analyzed to identify candidate eggs for diagnosis and counting”; page 7, section 3.2: “After the combined neural network image segmentation and morphological discrimination, final counts of numbers of eggs per sample are obtained”), and an infection status of the at least one parasite species; (Cumming, ¶ [0065]: “The present invention is predicated, in one aspect, at least in part on the finding that machine learning techniques may be used to identify the presence or absence of target biological material”)
The combination of Li, Cumming and Suwannaphong does not explicitly disclose displaying the output.
Pollak is in the same field of art of detecting parasitic infection in biological sample. Further, Pollak teaches displaying the output. (Pollak, ¶ [0159-0160]: “The processing unit is also configured to communicate with an output unit. Thus, based on the output value or values of the structural features determined by the processing unit, an output signal or output signals indicative of presence or absence of at least one pathogen and in some embodiments, the type of the pathogen, in the acquired image is provided by an output unit … including a graph, graphic or text displayed on a monitor of a control unit, a printout, as a voice message, or a user's smartphone display”)
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Li, Cumming and Suwannaphong by incorporating outputting unit that is taught by Pollak, to make a system to detect parasitic infection and sharing the result with user; thus, one of ordinary skilled in the art would be motivated to combine the references since among its several aspects, the present invention recognizes there is a need to improve user experience in such system by sharing result with user (Pollak, ¶ [0159-0160]).
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
CLAIM 17
Regarding Claim 17, the combination of Li, Cumming, Suwannaphong and Pollak teaches the medium of Claim 15. In addition, the combination of Li, Cumming, Suwannaphong and Pollak teaches the infection status comprises a binary status of an infected status or a non-infected status. (Cumming, ¶ [0065]: “The present invention is predicated, in one aspect, at least in part on the finding that machine learning techniques may be used to identify the presence or absence of target biological material”)
CLAIM 19
Regarding Claim 19, the combination of Li, Cumming, Suwannaphong and Pollak teaches the medium of Claim 15. In addition, the combination of Li, Cumming, Suwannaphong and Pollak teaches the operation of reconstructing the evaluated on-site dataset comprises remerging portions of cropped oocytes (Suwannaphong, page 2, section 2.2.1: “Each microscopic image is divided into patches, allowing the model to characterise the whole image by analysing local areas … merging probability of all patches together to reconstruct the probability map corresponding the input microscopic image…” Suwannaphong teaches dividing an original input image to patches, analyzing each patch for parasite eggs, then merge all patches together to reconstruct the a probability map corresponding to the original image) and correcting counts of oocytes for at least one parasite species. (Li, page 4, section 2.2.1: “Once images have been acquired, they must be automatically analyzed to identify candidate eggs for diagnosis and counting”; page 7, section 3.2: “After the combined neural network image segmentation and morphological discrimination, final counts of numbers of eggs per sample are obtained”)
CLAIM 20
Regarding Claim 20, the combination of Li, Cumming, Suwannaphong and Pollak teaches the medium of Claim 15. In addition, the combination of Li, Cumming, Suwannaphong and Pollak teaches the operation of performing the at least one cropping comprises performing the at least one cropping to allow images in the expanded on-site dataset to share a similar resolution to a resolution of images in an expanded training dataset. (Cumming, ¶ [0115]: “The size of the frame may be pre-determined according to an expected ovum size”; ¶ [0118-0121]: “A window of fixed dimension (width and height) is used to slide through the entire input image … At each position of the window on the input image, an image frame, corresponding to the current window, is cropped out ... Window size may be selected according to the size of an expected parasite ovum…” Cumming teaches resolution/size of cropped area is predetermined by expected parasite eggs, thus cropping on on-site images and cropping on training dataset of same expected parasite eggs will share the same resolution)
Claim(s) 4 and 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Li in view of Cumming in view of Suwannaphong in view of Pollak, and further in view of Nguyen et al. (US-20240029896-A1, filed 2021, hereinafter Nguyen).
CLAIM 4
In regards to Claim 4, the combination of Li, Cumming, Suwannaphong and Pollak teaches the device of Claim 3.
The combination of Li, Cumming, Suwannaphong and Pollak does not explicitly disclose storing aspects of the output in a central database to supplement a training database.
Nguyen is in the same field of art of training machine learning model. Further, Nguyen teaches storing aspects of the output in a central database to supplement a training database. (Nguyen, ¶ [0050]: “The neural network system can also be coupled to a training database ... The training database may provide a large sample of images used to train the neural networks, ... Communication between the training database and the neural network system can be bidirectional, such that the training database may provide images to the neural networks for training purposes, and the neural networks can transmit images for storage in the training database, thereby increasing the image sample size and further refining future output from the neural networks.” Nguyen teaches transmitting output from neural networks to a database for future training)
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Li, Cumming, Suwannaphong and Pollak by incorporating the method of expanding training database by output from machine learning models that is taught by Nguyen, to make machine learning system that can generate training data; thus, one of ordinary skilled in the art would be motivated to combine the references since among its several aspects, the present invention recognizes there is a need to improve effectiveness of training machine learning models by increase training database volume (Nguyen, ¶ [0050]: “ … increasing the image sample size and further refining future output from the neural networks”).
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
CLAIM 16
In regards to Claim 16, the combination of Li, Cumming, Suwannaphong and Pollak teaches the medium of Claim 15.
The combination of Li, Cumming, Suwannaphong and Pollak does not explicitly disclose storing aspects of the output in a central database to supplement a training database.
Nguyen is in the same field of art of training machine learning model. Further, Nguyen teaches storing aspects of the output in a central database to supplement a training database. (Nguyen, ¶ [0050]: “The neural network system can also be coupled to a training database ... The training database may provide a large sample of images used to train the neural networks, ... Communication between the training database and the neural network system can be bidirectional, such that the training database may provide images to the neural networks for training purposes, and the neural networks can transmit images for storage in the training database, thereby increasing the image sample size and further refining future output from the neural networks.” Nguyen teaches transmitting output from neural networks to a database for future training)
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Li, Cumming, Suwannaphong and Pollak by incorporating the method of expanding training database by output from machine learning models that is taught by Nguyen, to make machine learning system that can generate training data; thus, one of ordinary skilled in the art would be motivated to combine the references since among its several aspects, the present invention recognizes there is a need to improve effectiveness of training machine learning models by increase training database volume (Nguyen, ¶ [0050]: “ … increasing the image sample size and further refining future output from the neural networks”).
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
CLAIM 11
Claim(s) 11 is/are rejected under 35 U.S.C. 103 as being unpatentable over Li in view of Cumming, and further in view of Huang et al. (Huang, Pengchi, and Yongjian Zhu. "Multi-task data augmentation method joint object detection and semantic segmentation." IEEE, published 2022, hereinafter Huang).
Regarding Claim 11, the combination of Li and Cumming teaches the method of Claim 8.
The combination of Li and Cumming does not explicitly disclose performing at least one cropping or at least one rotation comprises applying a comparable transformation to segmentation labels to preserve ground-truth labels for training the machine-learning model.
Huang is in the same field of art of training machine learning model for detection task. Further, Huang teaches performing at least one cropping or at least one rotation comprises applying a comparable transformation to segmentation labels to preserve ground-truth labels for training the machine-learning model. (Huang, page 135, section II, subsection A: “… The data augmentation methods commonly used in target detection tasks include cropping, rotation, flipping, color distortion, CopyPaste [9], etc. Since the label format in the object detection task is coordinates, after performing morphological transformation on the original image, the same transformation needs to be performed on the corresponding label coordinates in the image.”)
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Li and Cumming by incorporating data augmentation method that is taught by Huang, to make a system that can expand training dataset by image augmentation; thus, one of ordinary skilled in the art would be motivated to combine the references since among its several aspects, the present invention recognizes there is a need to improve performance of machine learning model (Huang, page 134, section 1, last paragraph: “data augmentation is a work that can effectively improve the performance of neural networks … so these methods can be used in the multi-task training of automatic driving to enhance the data at the same time and improve the model performance”).
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
CLAIM 12
Claim(s) 12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Li in view of Cumming, and further in view of Cheng et al. (CN-111539957-A, published 2020, hereinafter Cheng).
Regarding Claim 12, the combination of Li and Cumming teaches the method of claim 8.
The combination of Li and Cumming does not explicitly disclose performing the at least one rotation comprises performing multiple rotations of each image in the dataset of labeled images at angular increments between angular boundaries.
Cheng is in the same field of art of training machine learning model for detection task. Further, Cheng teaches performing the at least one rotation comprises performing multiple rotations of each image in the dataset of labeled images at angular increments between angular boundaries. (Cheng, ¶ [0042]: “The data augmentation method includes geometric transformation operations and/or pixel transformation operations; preferably, the geometric transformation operations include or one more of rotation operations, scaling operations, and cropping operations… The rotation operation involves rotating the image clockwise or counterclockwise by a certain angle to reduce the probability of recognition failure when the image is tilted”; ¶ [0038]: “. In the data augmentation method, if rotation is performed, the rotation angle is m*90° (m can be selected from 0, 1, 2, 3)”. Cheng teaches multiple rotations of 0, 90, 180 and 270 degree)
Therefore, it would have been obvious to one having ordinary skill in the art before the effective filing date of the claimed invention to modify the invention of Li and Cumming by incorporating data augmentation method that is taught by Cheng, to make a system that can expand training dataset by image augmentation; thus, one of ordinary skilled in the art would be motivated to combine the references since among its several aspects, the present invention recognizes there is a need to improve performance of machine learning model (Cheng, ¶ [0007]: “… providing an image sample generation method, system and detection method for target detection … to improve the detection efficiency and accuracy in the target detection process using deep learning methods.” ).
Thus, the claimed subject matter would have been obvious to a person having ordinary skill in the art before the effective filing date of the claimed invention.
Allowable Subject Matter
Claims 6, 9-10 and 18 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
And
Pertinent Arts
The prior art made of record and not relied upon is considered pertinent to applicant’s disclosure.
Butploy et al. (Butploy, Narut, Wanida Kanarkard, and Pewpan Maleewong Intapan. "Deep learning approach for Ascaris lumbricoides parasite egg classification." Journal of Parasitology Research, published 2021, hereinafter Butploy), which is directed to detecting parasitic eggs, and classifying detected eggs into 3 categories: infertile, fertile and decorticate.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to NHUT HUY (JEREMY) PHAM whose telephone number is (703)756-5797. The examiner can normally be reached Mo - Fr. 8:30am - 6pm ET.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, O'Neal Mistry can be reached on (313)446-4912. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of 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.
/NHUT HUY PHAM/Examiner, Art Unit 2674 /Ross Varndell/Primary Examiner, Art Unit 2674