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 .
Status of the Claims
Applicant’s election without traverse of (Species A) claims 1-2, 13-14, 16-17, 26, 28, and 30 in the reply filed on April 9, 2026, was acknowledged in the prior office action.
Claims 3-12, 15, 17-25, 27, 29, 31-32, 34-41, 43, 45, 47, 49-52, 55-62, 64, 67-68, 72, 76-85, 87, 89, 91, 93-100, and 102-108 are cancelled.
Claims 33, 42, 44, 46, 48, 53-54, 63, 65-66, 69-71, 73-75, 86, 88, 90, 92, and 101 are withdrawn as being directed to a non-elected invention.
Claims 1, 13, 14, and 16 are amended and claim 17 is cancelled by the amendment filed August 19, 2026.
Claims 1-2, 13-14, 16, 26, 28 and 30 are pending and have been examined.
Response to Arguments
This final office action is in response to the amendment filed August 19, 2026. Regarding the 35 USC § 101, applicants arguments have been considered but they are not persuasive. Applicant argues that claim 1 does not recite a mathematical concept, relying on Example 39; however, the claim is more line Example 47, claim 2, where the backpropagation and gradient descent algorithms were found to be mathematical calculations although no formula was recited. The rejection has been updated to address the amended claims, as set forth below.
Applicant’s arguments with respect to claims 1, 2, 13, 14, 16, 26, 28, and 30, regarding the 35 USC § 102 and 103 rejections, have been considered but are moot in view of new ground(s) of rejection because of the amendments. The rejection under 35 USC § 102 over Yip is withdrawn, and the claims are now rejected under 35 USC § 103 over Yip in view of Yoo, as set forth below.
Regarding the double patenting, applicant’s arguments have been considered and are persuasive. The rejection of the claims under nonstatutory double patenting over copending application 19/391554 is withdrawn.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-2, 13-14, 16, 26, 28, and 30 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The claim(s) recite(s) mental processes and a mathematical concept. This judicial exception is not integrated into a practical application because the additional elements are generic data gathering, i.e., insignificant pre-solution activity, and a generic computer and generic machine learning models that merely apply the exception. The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception because receiving a tumor image is well-understood, routine, and conventional in digital pathology, and the computer and machine learning models are mere instructions to apply the exception.
Step 1 (Statutory Category): Claim 1 recites a method for determining a tumor immunophenotype and therefore falls within the statutory category of a process.
Step 2A, Prong One (Judicial Exception): The claim recites “calculating an epithelium-immune cell density for the image based on a number of immune cells detected within the one or more regions of the image depicting the tumor epithelium,” which is a mathematical calculation; specifically, a computation of a ratio or count of detected cells within identified regions.
The claim further recites, “determining at least a first density threshold by ranking a plurality of epithelium-immune cell densities corresponding to a plurality of reference images, each reference image depicting tumors having pre-determined tumor immunophenotypes,” which is a mathematical concept: ordering a set of density values and taking a cutoff from the ordered set. The specification describes the images being ranked based on epithelium-immune cell density (¶ 113), and the threshold is set from the densities of the images in the top 40%, next 40%, and bottom 20% of the ranking (¶¶ 114-115).
The claim further recites “determining a tumor immunophenotype of the image based on the epithelium-immune cell density and the at least first density threshold for classifying images into one of a set of tumor immunophenotypes,” which is a mathematical comparison of a computed value against a threshold value to perform a classification, and also constitutes a mental process in that a pathologist could mentally compare a density value to a threshold and render a classification judgment.
The claim also recites “identifying, by one or more machine learning models, one or more regions of the image depicting tumor epithelium,” which, apart from the machine learning models addressed as additional elements below, is a mental process – an observation and evaluation that a human pathologist routinely performs when examining a stained tissue image. The specification confirms that pathologists have traditionally performed this type of analysis manually. See ¶¶3-4 (acknowledging that manually categorizing tumors based on the spatial distribution of immune effector cells is predictive for checkpoint inhibitor therapies). These limitations, considered together, constitute a single abstract idea encompassing both mathematical concepts and mental processes. MPEP 2106.04(a)(2)(I), (III).
Step 2A Prong Two (Practical Application): The claim does not integrate the judicial exception into a practical application.
The additional elements “computer implemented” and “by one or more machine learning models” are recited at a high level of generality, with no explanation of how the models identify the epithelium regions, and amount to mere instruction to apply the exception (MPEP 2106.05(f)). The claim applies generic machine learning to a new environment. Claims that “do no more than claim the application of generic machine learning to new data environments, without disclosing improvements to the machine learning models to be applied” are not eligible. Recentive Analytics, Inc. v. Fox Corp., 134 F.4th 1205 (Fed. Cir. 2025). By contrast, claims directed to improving the functioning of the model itself may be eligible. MPEP 2106.04(d), subsection III, and MPEP 2106.05(a), citing Ex parte Desjardins. No such improvement to the model is recited here; the claim recites only that the models identify epithelium regions, with no architecture, training, or operational detail. The additional element of “receiving an image of a tumor” is insignificant extra-solution activity in the form of mere data gathering, recited at a high level of generality i.e., insignificant pre-solution activity (MPEP 2106.05(g)). The claim does not recite any particular imaging apparatus, any specific treatment or prophylaxis for a patient based on the determined immunophenotype, any transformation of a particular article, or any improvement to the functioning of a computer or other technology. The claim terminates at the step of determining a tumor immunophenotype – a classification result without requiring any real-world application of that result. The claim as a whole merely applies known analytical techniques (density calculation and threshold-based classification) to a particular technological environment (digital pathology images), which amounts to generally linking the judicial exception to a field of use. See MPEP §2106.05(h). The claim does not reflect
the asserted improvement of standardizing the tumor immunophenotyping process because it does not recite the specific technical components or steps that achieve such improvement; rather, it recites the abstract analytical steps themselves at a high level of generality. Nor does the claim reflect the asserted reduction in computational resources; it recites no computing resource or processing constraint.
Step 2B (Inventive Concept): The claim does not recite additional elements that amount to
significantly more than the judicial exception. The limitation of "receiving an image of a tumor" was identified as insignificant extra-solution activity at Step 2A, Prong Two and is reevaluated
here. This limitation amounts to well-understood , routine, and conventional activity. See, e.g., Chukka US 2016/0042511 ¶¶ 4, 27, 28 (routine imaging apparatus capture of digital whole-slide images).
Receiving digital pathology images for analysis is conventional data gathering in the field of computational pathology. The computer and machine learning models are mere instructions to apply the exception and do not provide an inventive concepts (MPEP 2106.05(f)). The remaining limitations constitute the abstract idea itself and do not provide an inventive concept. Accordingly, claim 1 is not patent eligible.
Dependent Claims:
Claim 2 further recites mental-process steps (sliding-window scan and classify region as epithelium or stroma). Same analysis. Ineligible.
Claim 13 recites "using the one or more machine learning models" at high level of generality. This is a mere instruction to apply the exception under MPEP 2106.05(f). Generic ML for a pathological counting task previously performed by hand is well-understood, routine, and conventional. See Yip US 2023/0230195 A1 ¶9 (H&E "is a long-standing method used by pathologists to analyze tissue morphological features"); Chukka ¶¶ 22, 23 (training a classifier on labeled ground truth slides). Same analysis. Ineligible.
Claim 14 recites that the machine learning models of claim 1 "comprise a computer vision model trained to recognize immune cells." This is generic computer vision invoked at high level of generality. It is a mere instruction to apply the abstract idea on a conventional ML tool. MPEP 2106.05(f). Under Step 2B, a computer vision model trained to recognize lymphocytes
(i.e., immune cells) is well-understood, routine, and conventional in digital pathology. See Yip
US 2023/0230195 A1 ¶120 ("the model 316 may be configured to perform lymphocyte identification and segmentation using a three-class FCN model"); Chukka US 2016/0042511 A1 ¶23 (ground-truth training data identifies “a lymphocyte” as one of the object types); Chukka ¶46 (classifier outputs include “a lymphocyte”). Ineligible.
Claim 16 recites that each reference image “is associated with a patient of a plurality of patients participating in a first clinical trial” and “includes a label indicating the pre-determined tumor immunophenotype of the image.” These limitations identify the source and labels of the data the ranking step uses, with is data gathering and field of use. (MPEP 2106.05(g), (h)). Same analysis. Ineligible.
Claim 26 recites "training a classifier" at high level of generality. Same MPEP 2106.05(f) defect as claim 13. Ineligible.
Claim 28 recites "training one or more machine learning models to detect biological objects" using "training data comprising a plurality of images and labels indicating a type of biological object." Supervised training of an ML model on labeled image data is generic ML invoked at high level of generality, disclosed in MPEP 2106.05(f) "apply it" defect. Under Step 2B, supervised pixel-level training of cell-segmentation models on labeled pathology images is well-understood, routine, and conventional. See Yip ¶124 ("training data includes digital slide images where every pixel has been labeled as either the interior of a cell, the outer edge of a cell, or the background"); Chukka ¶22 (training "an object classifier. .. using a plurality of 'ground truth' sample slides or training images"). Ineligible.
Claim 30 recites that the image "comprises a digital pathology image captured using a digital pathology imaging system." Recitation of a generic "digital pathology imaging system" does not invoke a particular machine under MPEP 2106.05(b). The recitation refers to the class of standard whole-slide scanners, not any specific machine with meaningful limits on the claim. Image capture by such a system is also insignificant pre-solution activity under MPEP 2106.05(g). Under Step 2B, digital whole-slide image acquisition is well-understood, routine, and conventional. See Yip ¶84 ("any suitable optical histopathology slide scanner including 20x and 40x resolution magnification scanners"); Chukka ¶4 (routine "digital whole-slide image" acquisition); Chukka ¶28 (standard imaging apparatus components such as a camera and CCD sensor). Ineligible.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1, 2, 13, 14, 16, 26, 28, and 30 is/are rejected under 35 U.S.C. 103 as being unpatentable over Yip et al., US 2023/0230195 A1 (hereinafter “Yip”) in view of Yoo et al., US 2022/0036971 A1 (hereinafter “Yoo”).
Claim 1.
Yip in view of Yoo discloses a computer-implemented method for determining a tumor immunophenotype (Yip: “The imaging-based biomarker prediction system 102 may be implemented on one or more computing device, such as a computer, tablet or other mobile computing device, or server, such as a cloud server” (¶82).), comprising:
determining at least a first density threshold (Yip: “By labeling the image data 162a according to associations with the other data types, the imaging based biomarker prediction system may train an image classifier module to predict the one or more different data types from image data 162a” (¶93); the data types “the immune state a tumor” (¶ 97); and “The data types 162a-162d may be used by the histopathology image-based classifier training module 160 to develop classifiers for identifying one of more of the biomarkers discussed herein” (¶ 98). This teaches reference images trained with a pre-determined tumor immune state. Yip compares a lymphocyte density against a threshold: “A second layer may divide the lymphocyte cell count by the cell count to generate a ratio, which when compared to a threshold may be used to assign the biomarker status in the image based on whether the ratio exceeds the threshold” (¶ 165). Determining the threshold by ranking the densities is taught by Yoo, as set forth below.);
receiving an image of a tumor (YIP: "Hematoxylin and eosin (H&E) staining is a long-standing method used by pathologists to analyze tissue morphological features for malignancy diagnosis. H&E slides, for example, can illustrate visual characteristics of tissue structures such as cell nuclei and cytoplasm, to inform identification of cancer tumors" (¶ 9); "the
imaging-based biomarker prediction system 102 is communicatively coupled to receive
medical images, for example of histopathology slides such as digital H&E stained slide
images"(¶ 84). This teaches receiving a digital H&E image of a tumor.);
identifying, by one or more machine learning models, one or more regions of the image depicting tumor epithelium (YIP: “deep learning framework 306 may be configured using
a FCN classifier in an example” (¶ 136); and "the tissue classification model 320 may classify tissue using tissue classifications, such as Tumor-IHC positive, Tumor-IHC negative, Necrosis, Stroma, Epithelium, or Blood"(¶ 137). This teaches a trained FCN classifier identifying regions of the image as the "Epithelium" tissue class.);
calculating an epithelium-immune cell density for the image based on a number of immune cells detected within the one or more regions of the image depicting the tumor epithelium (YIP: "Information regarding TIL density, location, organization, and composition provide valuable insight as to prognosis and potential treatment options. In various aspects, the disclosure provides a method of predicting TIL density in a sample, a method of distinguishing subpopulations of TILs in a sample ... , a method of distinguishing stromal versus intratumoral TILs, and the like" (¶ 67); "the tissue classification model 320 is trained to identify a percentage TILs within a tile image, the cell segmenter 316 determines the cell boundary, and the biomarker classification model 322 classifies the tile image based on the percentage of TILs within a cell interior" (¶ 130); and "the system 300 then takes this new list of locations of
confirmed lymphocytes from the module 304 and compares to a tissue segmenter module
318 list of tissue, e.g., tumor and non-tumor tissue locations determined from tissue classification model 320 and determines whether the lymphocyte is in tumor or non-tumor
region" (¶ 121). This teaches TIL density quantified per region, distinguishing
intratumoral (i.e., within the epithelium compartment) from stromal TILs.); and
determining a tumor immunophenotype of the image based on the epithelium-immune cell density and the at least first density threshold for classifying images into one of a set of tumor immunophenotypes (YIP: the lymphocyte region, “when compared to a threshold may be used to assign the biomarker status in the image based on whether the ratio exceeds the
threshold” (¶ 165); and “the immune state of a tumor (for example, inflamed/'hot' vs. non-inflamed/'cold' vs immune excluded)” (¶ 97). This teaches assigning a categorical tumor immunophenotype (inflamed/non-inflamed/immune excluded) based on whether the lymphocyte density exceeds a threshold.).
Yip discloses all of the subject matter as described above except for specifically teaching determining the first density threshold “by ranking” the plurality of epithelium-immune cell densities. However, Yoo in the same field of endeavor teaches the ranking (Yoo: “FIG. 4, if the density of the immune cells in the cancer area is equal to or greater than a first threshold density, the immune phenotype of one or more regions of interest may be determined to be immune inflamed” (¶ 103); “the first threshold density may correspond to a density value corresponding to top X % (where X is a number between 0 and 100) in the distribution for the density of immune cells in the cancer area in each of the regions of interest of the plurality of pathology slide images” (¶ 104); and “the cancer area may be used interchangeably with cancer epithelium” (¶ 71). This teaches setting the epithelium-immune cell density threshold at the density value at a rank position, the top X %, among the densities ordered across a plurality of pathology slide images, each density corresponding to the slide image from which its region of interest was taken.).
It would have been obvious to one of ordinary skill before the effective filing date of the claimed invention to set Yip’s density threshold (¶ 165) the way Yoo sets its own, at the top X % of the ranked densities (Yoo ¶ 104), using Yip’s labeled images as the reference set, because “objectively analyzing the immune environment around the cancer cells,” improves “the predictive rate” (¶ 27), and a cutoff taken from ranked densities is objective. This is the use of a known technique to improve similar devices in the same way (MPEP 2143(C)).
Claim 2.
Yip in view of Yoo discloses the method of claim 1, wherein identifying the one or more regions comprises: scanning the image using a sliding window; and for each portion of the image included within the sliding window: classifying the portion as at least one of a region depicting tumor epithelium or a region depicting tumor stroma (YIP: "image data is to be analyzed on a tile-basis, in some examples, image pre-processing includes receiving an initial histopathology image, at a first image resolution, downsampling that image to a second image resolution, and then performing a normalization on the downsampled histopathology image" (¶ 87); and "the deep learning framework module 306 is configured to classify tissue using a tiling analysis. For example, in the pipeline 315, the tissue detection process sends histopathology images ... to the image tiling process that selects and applies a tiling mask to the received images to
parse the images into small sub-images" (¶ 115). The tile classifier outputs "Tumor-IHC
positive, Tumor-IHC negative, Necrosis, Stroma, Epithelium, or Blood" (¶ 137).).
Claim 13.
Yip in view of Yoo discloses the method of claim 1, wherein calculating the epithelium-immune cell density comprises: determining the number of immune cells detected within each of the one or more regions of the image using one or more machine learning models (YIP: "the tissue classification model 320 is trained to identify a percentage TILs within a tile image, the cell segmenter 316 determines the cell boundary, and the biomarker classification model 322 classifies the tile image based on the percentage of TILs within a cell interior" (¶ 130); the framework is "a deep learning framework" (¶ 53). This teaches using trained ML models for the TIL-counting step.).
Claim 14.
Yip in view of Yoo discloses the method of claim 13, wherein the one or more machine learning models comprise a computer vision model trained to recognize immune cells (YIP: "the cell segmentation model 316 of the module 304 may be configured as a three-class semantic segmentation FCN model developed by modifying a UNet classifier"(¶ 120); "For both TILs biomarkers, the model 316 may be configured to perform lymphocyte identification and segmentation using a three-class FCN model" (¶ 120). This teaches a computer vision model (UNet/FCN) trained to recognize lymphocytes, i.e., immune cells.).
Claim 26.
Yip in view of Yoo discloses the method of claim 1, further comprising: training a classifier to determine a tumor immunophenotype of an image input to the classifier based on a calculated epithelium-immune cell density of the image input to the classifier, the first density threshold, and a second density threshold (Yoo: “the first threshold density or the second threshold density may include a threshold density for an upper area and a threshold density for a lower area” (¶ 105); “the artificial neural network model for immune phenotype classification may
correspond to a model that is trained to determine at least one of an immune phenotype … may include at least one of: a statistical feature for one or more target items” (¶ 74); This teaches training a classifier to determine the immune cell density using an upper and lower threshold. The reason to combine is the same as for claim 1.).
Claim 28.
The method of claim 1, further comprising: training one or more machine learning models to detect biological objects, wherein the one or more machine learning models are trained using training data comprising a plurality of images and labels indicating a type of biological object depicted within each of the plurality of images, wherein the epithelium-immune cell density is calculated based on the one or more machine learning models (YIP: "training data includes digital slide images where every pixel has been labeled as either the interior of a cell, the outer edge of a cell, or the background which is exterior to every cell" (¶ 124); "the deep learning framework 150 includes a histopathology image based classifier training module 160 that can access received and stored data from the external systems"(¶ 93); and Yip's claim 17 recites "receive a plurality of digital images of Hematoxylin and Eosin-stained training slides of training tissue samples corresponding to the respective different biomarker ... and generate one of the trained biomarker classification models, based on the plurality of digital images." This teaches supervised training using labeled images.).
Claim 30.
The method of claim 1, wherein the image comprises a digital pathology image captured using a digital pathology imaging system (YIP: "the imaging-based biomarker prediction system 102 is communicatively coupled to receive medical images ... from a variety of different sources. These sources may include a physician clinical records system 106 and a histopathology imaging system 108. Any number of medical image data sources could be accessible using the system 100. The histopathology images may be images captured by any dedicated digital medical image scanners, including any suitable optical histopathology slide scanner including 20x and 40x resolution magnification scanners" (¶ 84). This teaches a digital pathology imaging system.).
Claim 16.
Yip in view of Yoo discloses the method of claim 1, wherein: each reference image of the plurality of reference images is associated with a patient of a plurality of patients and each reference image of the plurality of reference images includes a label indicating the predetermined tumor immunophenotype of the image (YIP: “An association may be formed by labeling the image data 162a with one or more of the different data types” (¶ 93), the data types including “the immune state of a tumor” (¶ 97); “images and other data of previously treated and analyzed patients is utilized, through the trained models, to provide analysis and diagnosis capabilities for future patients” (¶ 92); tumor response data “including data about a reduction in the growth of the tumor after exposure to certain therapies, for example immunotherapies” (¶ 96); and “The report may further include clinical trials for which the patient is eligible” (¶ 154). This teaches reference images of treated patients, each labeled with the immune state of the tumor.).
Yip in view of Yoo does not expressly teach that the patients are "participating in a first clinical trial.” It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to draw Yip's labeled reference images from patients in a clinical trial, because Yip already trains on “previously treated and analyzed patients” (Yip ¶ 92) and ties its patients to “clinical trials for which the patient is eligible" (Yip ¶ 154). This is combining prior art elements according to known methods to yield predictable results (MPEP 2143(A)). Yip has the labeled images and the trial link (¶¶ 92, 97, 154). A trial cohort changes where the images come from, not any method step. The images still supply the densities the threshold is taken from. The result is predictable.
Conclusion
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Ross Varndell whose telephone number is (571)270-1922. The examiner can normally be reached M-F, 9-5 EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, O’Neal Mistry can be reached at (313)446-4912. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see https://ppair-my.uspto.gov/pair/PrivatePair. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/Ross Varndell/Primary Examiner, Art Unit 2674
Prosecution Insights