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 .
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 08/28/2024 has been considered by the examiner.
Status of Claims
Claims 1-20 are currently pending in this application.
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 13 and 14 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 13 recites the limitation "a corresponding one of the LSTM cells" in lines 3. There is insufficient antecedent basis for this limitation. It is unclear what the “LSTM cells” is referring to, is it referring to “the LSTM network”? For purposes of examination , the limitation will be interpreted as “the LSTM network”. Further, “a corresponding one of the LSTM cells” lacks antecedent basis as claim 12 introduces a singular LSTM network. Examiner suggest removing “a corresponding one of…”.
Claim 14 is rejected due to its dependency.
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.
Claim(s) 1-2, 8-11, 17, and 19-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The limitations, under their broadest reasonable interpretation, cover mental process (concept performed in a human mind, including as observation, evaluation, judgment, opinion). The independent claims 1, 10, and 19 recite an apparatus, a non-transitory computer readable medium and a method. This judicial exception is not integrated into a practical application because the steps do not add meaningful limitations to be considered specifically applied to a particular technological problem to be solved .The claim(s) does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the steps of the claimed invention can be done mentally and no additional features in the claims would preclude them from being performed as such except for the generic computer elements at high level of generality (i.e., processor, memory).
According to the USPTO guidelines, a claim is directed to non-statutory subject matter if:
STEP 1: the claim does not fall within one of the four statutory categories of invention (process, machine, manufacture or composition of matter), or
STEP 2: the claim recites a judicial exception, e.g. an abstract idea, without reciting additional elements that amount to significantly more than the judicial exception, as determined using the following analysis:
STEP 2A (PRONG 1): Does the claim recite an abstract idea, law of nature, or natural phenomenon?
STEP 2A (PRONG 2): Does the claim recite additional elements that integrate the judicial exception into a practical application?
STEP 2B: Does the claim recite additional elements that amount to significantly more than the judicial exception?
Using the two-step inquiry, it is clear that the independent claims 1, 10, and 19 are directed to an abstract idea as shown below:
STEP 1: Do the claims fall within one of the statutory categories? YES. Independent claims 1, 10 , and 19 are directed to a machine, manufacture, and process.
STEP 2A (PRONG 1): Is the claim directed to a law of nature, a natural phenomenon or an abstract idea? YES, the claims are directed toward a mental process (i.e. abstract idea).
With regard to STEP 2A (PRONG 1), the guidelines provide three groupings of subject matter that are considered abstract ideas:
Mathematical concepts – mathematical relationships, mathematical formulas or equations, mathematical calculations;
Certain methods of organizing human activity – fundamental economic principles or practices (including hedging, insurance, mitigating risk); commercial or legal interactions (including agreements in the form of contracts; legal obligations; advertising, marketing or sales activities or behaviors; business relations); managing personal behavior or relationships or interactions between people (including social activities, teaching, and following rules or instructions); and
Mental processes – concepts that are practicably performed in the human mind (including an observation, evaluation, judgment, opinion).
Independent claims 1, 10, and 19 comprise a mental process that can be practicably performed in the human mind (or generic computers or components configured to perform the method) and, therefore, an abstract idea.
Regarding independent claim(s) 1: the limitations recite:
An apparatus for detecting a medical condition in a histopathology image (additional element), comprising:
a hardware memory configured to store executable instructions (generic computer component); and
a hardware processor in communication with the hardware memory, wherein the executable instructions, when executed by the processor, cause the processor to (generic computer component):
obtain a plurality of patches at a plurality of magnification levels from the histopathology image (data gathering), apply a deep learning algorithm to each of the patches (generic computer component),
extract, from applying the deep learning algorithm (generic computer component), information representative of a hierarchical relationship that links characteristics of the histopathology image present at one level and another level of the plurality of magnification levels (mental process including observation and evaluation, and can be done mentally in the human mind), and
identify the medical condition based on the extracted information representative of the hierarchical relationship for characteristics present at the one level and at the another level of the plurality of magnification levels (mental process including observation and evaluation, and can be done mentally in the human mind).
Regarding independent claim(s) 10: the limitations recite:
A non-transitory computer readable medium for detecting a medical condition in a histopathology image, the computer readable medium having program instructions for causing a hardware processor to (generic computer component):
obtain a plurality of patches at a plurality of magnification levels from the histopathology image (data gathering); apply a deep learning algorithm to each of the patches (generic computer component);
extract, from applying the deep learning algorithm (generic computer component), information representative of a hierarchical relationship that links characteristics of the histopathology image present at one level and another level the plurality of magnification levels (mental process including observation and evaluation, and can be done mentally in the human mind); and
identify the medical condition based on the extracted information representative of the hierarchical relationship for characteristics present at the one level and the another level of the plurality of magnification levels (mental process including observation and evaluation, and can be done mentally in the human mind).
Regarding independent claim(s) 19: the limitations recite:
A method, comprising:
obtaining a plurality of patches at a plurality of magnification levels from a histopathology image (data gathering); applying a deep learning algorithm to each of the patches (generic computer component);
extracting, from applying the deep learning algorithm (generic computer component), information representative of a hierarchical relationship that links characteristics of the histopathology image present at one level and another level of the plurality of magnification levels (mental process including observation and evaluation, and can be done mentally in the human mind); and
identifying a medical condition based on the extracted information representative of the hierarchical relationship for characteristics present at the one level and the another level of the plurality of magnification levels (mental process including observation and evaluation, and can be done mentally in the human mind).
These limitations, as drafted, is a simple process that, under their broadest reasonable interpretation, covers performance of the limitations in the mind or by a human. The Examiner notes that under MPEP 2106.04(a)(2)(III), the courts consider a mental process (thinking) that “can be performed in the human mind, or by a human using a pen and paper" to be an abstract idea. CyberSource Corp. v. Retail Decisions, Inc., 654 F.3d 1366, 1372, 99 USPQ2d 1690, 1695 (Fed. Cir. 2011). As the Federal Circuit explained, "methods which can be performed mentally, or which are the equivalent of human mental work, are unpatentable abstract ideas the ‘basic tools of scientific and technological work’ that are open to all.’" 654 F.3d at 1371, 99 USPQ2d at 1694 (citing Gottschalk v. Benson, 409 U.S. 63, 175 USPQ 673 (1972)). See also Mayo Collaborative Servs. v. Prometheus Labs. Inc., 566 U.S. 66, 71, 101 USPQ2d 1961, 1965 ("‘[M]ental processes[] and abstract intellectual concepts are not patentable, as they are the basic tools of scientific and technological work’" (quoting Benson, 409 U.S. at 67, 175 USPQ at 675)); Parker v. Flook, 437 U.S. 584, 589, 198 USPQ 193, 197 (1978) (same).
As such, a person could mentally evaluate characteristics of different patches at varying magnifications taken from a histopathology image, determine a hierarchical relationship among the characteristics at different magnification levels, and identify a medical condition based on the evaluated relationship. The mere nominal recitation that the various steps are being executed by the generic computer components, for example, a memory, processor, non-transitory computer readable medium, a deep learning algorithm, etc, does not take the limitations out of the mental process grouping. Thus, the claims recite a mental process.
STEP 2A (PRONG 2): Does the claim recite additional elements that integrate the judicial exception into a practical application? NO, the claims do not recite additional elements that integrate the judicial exception into a practical application.
With regard to STEP 2A (prong 2), whether the claim recites additional elements that integrate the judicial exception into a practical application, the guidelines provide the following exemplary considerations that are indicative that an additional element (or combination of elements) may have integrated the judicial exception into a practical application:
an additional element reflects an improvement in the functioning of a computer, or an improvement to other technology or technical field;
an additional element that applies or uses a judicial exception to affect a particular treatment or prophylaxis for a disease or medical condition;
an additional element implements a judicial exception with, or uses a judicial exception in conjunction with, a particular machine or manufacture that is integral to the claim;
an additional element effects a transformation or reduction of a particular article to a different state or thing; and
an additional element applies or uses the judicial exception in some other meaningful way beyond generally linking the use of the judicial exception to a particular technological environment, such that the claim as a whole is more than a drafting effort designed to monopolize the exception.
While the guidelines further state that the exemplary considerations are not an exhaustive list and that there may be other examples of integrating the exception into a practical application, the guidelines also list examples in which a judicial exception has not been integrated into a practical application:
an additional element merely recites the words “apply it” (or an equivalent) with the judicial exception, or merely includes instructions to implement an abstract idea on a computer, or merely uses a computer as a tool to perform an abstract idea;
an additional element adds insignificant extra-solution activity to the judicial exception; and
an additional element does no more than generally link the use of a judicial exception to a particular technological environment or field of use.
Independent claims 1, 10 , and 19 do not recite any of the exemplary considerations that are indicative of an abstract idea having been integrated into a practical application. Independent claims 1, 10, and 19 discloses an apparatus, memory, processor, non-transitory computer readable medium, obtaining a plurality of patches at a plurality of magnification levels from a histopathology image, and applying deep learning algorithm, which are generic computer components and/or insignificant pre/post-solution extra activity that do not add a meaningful limitation to the abstract idea because they amount to simply implementing the abstract idea in a method. Also, the claim invokes a generic deep learning algorithm merely as a tool for extracting information representative of a hierarchical relationship that links characteristics of the histopathology image rather than purporting to improve the technology or a computer. See MPEP 2106.05(f). Therefore, the limitation represents no more than mere instructions to apply the judicial exception on a computer. It can also be viewed as nothing more than an attempt to generally link the use of the judicial exception to the technological environment of computers.
These limitations are recited at a high level of generality (i.e. as a general action or change being taken based on the results of the acquiring step) and amounts to mere post solution actions, which is a form of insignificant extra-solution activity. Further, the claims are claimed generically and are operating in their ordinary capacity such that they do not use the judicial exception in a manner that imposes a meaningful limit on the judicial exception. Accordingly, even in combination, these additional elements do not integrate the abstract idea into a practical application because they do not impose any meaningful limits on practicing the abstract idea.
STEP 2B: Does the claim recite additional elements that amount to significantly more than the judicial exception? No, the claims do not recite additional elements that amount to significantly more than the judicial exception.
With regard to STEP 2B, whether the claims recite additional elements that provide significantly more than the recited judicial exception, the guidelines specify that the pre-guideline procedure is still in effect. Specifically, that examiners should continue to consider whether an additional element or combination of elements:
adds a specific limitation or combination of limitations that are not well-understood, routine, conventional activity in the field, which is indicative that an inventive concept may be present; or
simply appends well-understood, routine, conventional activities previously known to the industry, specified at a high level of generality, to the judicial exception, which is indicative that an inventive concept may not be present.
Independent claim(s) 1, 10, and 19 do not recite any additional elements that are not well-understood, routine or conventional. The use of a generic computer elements are routine, well-understood and conventional process that is performed by computers.
Thus, since independent claims 1, 10, and 19 are: (a) directed toward an abstract idea, (b) do not recite additional elements that integrate the judicial exception into a practical application, and (c) do not recite additional elements that amount to significantly more than the judicial exception, it is clear that independent claims 1, 10, and 19 are not eligible subject matter under 35 U.S.C 101.
Regarding claims 2, 8-9, 11, 17, and 20: the additional elements do no integrate the mental process into practical application or add significantly more to the mental process.
In detail claims 2 and 8-9 depend on claim 1, claims 11 and 17 depend on claim 10, and claim 20 depends on claim 19 and add:
cropping the histopathology image in a plurality of patches where in each patch has a different magnification (claims 2, 11, and 20).
wherein the hierarchical relationship represents a relation between tissue morphology at one magnification level and cell structure at another magnification level (claims 8 and 17).
crop the image to generate the plurality of patches based on a determined region of interest (claim 9).
Regarding claim 3: the additional limitations do integrate the mental process into practical application or add significantly more to the mental process. The limitation: "wherein the deep learning algorithm comprises a plurality of convolutional neural networks (CNNs) and a long-short term memory (LSTM) network, and wherein the executable instructions, when executed by the processor, further cause the processor to: provide each of the patches to a corresponding one of the CNNs, and provide an output of each of the CNNs to the LSTM network, wherein the LSTM network is configured to extract the information representative of the hierarchical relationship that links the characteristics of the histopathology image present at the one level and at the another level of the plurality of magnification levels." integrates the mental process into a practical application.
Regarding dependent claims 4-5 and 6-7: claims 4-5 and 6-7 are similarly eligible under 35 USC 101 due to their dependency on claim 3.
Regarding claim 12: the additional limitations do integrate the mental process into practical application or add significantly more to the mental process. The limitation: " wherein the deep learning algorithm comprises a plurality of convolutional neural networks (CNNs) and a long- short term memory (LSTM) network, and wherein the instructions are further configured to cause the hardware processor to: provide each of the patches to a corresponding one of the CNNs; and provide an output of each of the CNNs to the LSTM network, wherein the LSTM network is configured to extract the information representative of the hierarchical relationship that links the characteristics of the histopathology image present at the one level and at the another level of the plurality of magnification levels." integrates the mental process into a practical application.
Regarding dependent claims 13-14, 15-16, and 18: claims 13-14, 15-16, and 18 are similarly eligible under 35 USC 101 due to their dependency on claim 12.
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 following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-3, 6-7, 10-12, 15 -16, and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Stumpe et al. (US 11,170,897 B2) (hereinafter, “Stumpe”) in view of Zuo et al. ("Learning contextual dependence with convolutional hierarchical recurrent neural networks." IEEE Transactions on Image Processing 25.7 (2016): 2983-2996.) (hereinafter, “Zuo”).
Regarding claim 1, Stumpe discloses an apparatus for detecting a medical condition in a histopathology image (Abstract “A method, system and machine for assisting a pathologist in identifying the presence of tumor cells in lymph node tissue is disclosed. The digital image of lymph node tissue at a first magnification (e.g., 40×) is subdivided into a multitude of rectangular “patches.””), comprising:
a hardware memory configured to store executable instructions (Column 15 [lines 48-51] "the present disclosure are directed to systems, apparatus, tangible, non-transitory computer-readable media, user interfaces, memory devices, and electronic devices for analyzing a tissue biopsy."); and
a hardware processor in communication with the hardware memory, wherein the executable instructions, when executed by the processor (Column 15 [lines 48-51] "the present disclosure are directed to systems, apparatus, tangible, non-transitory computer-readable media, user interfaces, memory devices, and electronic devices for analyzing a tissue biopsy."), cause the processor to:
obtain a plurality of patches at a plurality of magnification levels (patch at 40x, 20x, 10x, 5x in Column 6 [lines 45-57] equates to plurality of patches) from the histopathology image, apply a deep learning algorithm (four members in the ensemble in Column 6 [lines 45-57] equate to the deep learning algorithm) to each of the patches (Column 6 [lines 45-57] “each member of an ensemble of deep neural network pattern recognizers analyze pixel information of the patch, including some surrounding pixels, and generates a probability score of between 0 and 1. Each member operates on a patch of pixels but at different magnification levels. Preferably, there are four such members in the ensemble. For example, one operates on a 128×128 patch at 40× magnification. Another operates on a 128×128 patch but at 20× magnification (centered on or containing the 128×128 patch at 40×). A third one operates on a 128×128 patch but at 10× (again, centered on or containing the 128×128 patch at 40×). A fourth one operates on a 128×128 patch at 5×"; Figure 5),
[extract, from applying the deep learning algorithm, information representative of a hierarchical relationship that links characteristics of the] histopathology image [present at one level and another level] of the plurality of magnification levels (Column 8 line 53 continuing to column 9 line 3 “Four different networks (306A, 306B, 306C, 306C) were trained corresponding to 5×, 10×, 20×, and 40× magnification…Each network 306A, 306B, 306C, 306D generates its own output 308A, 308B, 308C, 308D in the form of a score between 0 and 1.”), and
identify the medical condition based on the extracted information [representative of the hierarchical relationship] for characteristics present at the one level and at the another level of the plurality of magnification levels (Column 9 [lines 4-9] "The score between 0 and 1 is usually generated as the last layer of the neural network pattern recognizers 306A. 306B, 306C, 306D, in the form of a multinomial logistic regression, which generates a prediction, in the form of a probability of between 0 and 1, of which of the classes (here, healthy vs tumor) the input data (patch) belongs to."; Examiner interprets that the input data (each of the patches) correlates to the previously mentioned patches, in Column 6 [lines 45-57], at different magnifications (i.e. one level and another level)).
However, Stumpe fails to teach extract, from applying the deep learning algorithm, information representative of a hierarchical relationship that links characteristics of the [histopathology image] present at one level and another level [of the plurality of magnification levels].
Zuo teaches extract, from applying the deep learning algorithm (CNN in figure 1 equates to the deep learning algorithm), information representative of a hierarchical relationship that links characteristics of the [histopathology image] present at one level (lower level scales on Page 2988 equates to one level) and another level (the higher level scale Page 2988 equates to another level) [of the plurality of magnification levels] (Figure 1; Page 2988 last 2 paragraph of left column continuing to Right column first paragraph “local features are extracted from single-scale image regions. However, if the cross-scale information can be encoded, then better local descriptions can be achieved. Thus, we build recurrent connections across regions from different scales. For each element at each scale, its receptive field covers a number of elements at the lower level scales. More intuitively, as shown in the middle part of Figure 1, areas highlighted with yellow at the scale l+1 and l+2 are covered by the receptive field of the yellow element at the scale l . Thus, global information from the higher level scale l would be transferred to the corresponding areas at the lower level scales l+1 and l+2.”).
Therefore, it would have been obvious to one of ordinary skill of the art before the effective filing date to modify Stumpe’s reference to include extract, from applying the deep learning algorithm, information representative of a hierarchical relationship that links characteristics of the [histopathology image] present at one level and another level [of the plurality of magnification levels] taught by Zuo’s reference. The motivation for doing so would have been because dependencies among different image regions are very important for generating explicit image representations as suggested by Zuo (see Zuo, Abstract).
Further, one skilled in the art could have combined the elements described above by known methods with no change to the respective functions, and the combination would have yielded nothing more that predictable results. Therefore, it would have been obvious to combine Zuo with Stumpe to obtain the invention specified in claim 1.
Regarding claim 2, which claim 1 is incorporated, Stumpe discloses obtain the histopathology image (Column 5 [lines 27-34] "a biopsy of tissue potentially containing cancer cells (e.g., lymph node tissue) is obtained from a patient and is formalin fixed and paraffin embedded, optionally in a tissue block which is conventional. At step 102, the tissue block is sectioned, stained with H&E and laid on a clear glass slide. At step 104, the slide is digitally scanned by a whole slide digital slide scanner. At step 106, this scan produces a digitized image of the slide"), and
crop the histopathology image to generate the plurality of patches (Column 5 [lines 34-48] "The slide image is subdivided into a multitude of rectangular patches, such as ˜12,000 or so. The number of patches per slide can range from 10,000 to 400,000. In one embodiment each patch is in the form of a square of 128×128 pixels or 299×299 pixels."),
wherein each of the patches has a different magnification level from the other patches (Column 6 [lines 49-57] "The slide image is subdivided into a multitude of rectangular patches, such as ˜12,000 or so. The number of patches per slide can range from 10,000 to 400,000. In one embodiment each patch is in the form of a square of 128×128 pixels or 299×299 pixels.").
Regarding claim 3, which claim 1 is incorporated, Stumpe discloses wherein the deep learning algorithm comprises a plurality of convolutional neural networks (CNNs) [and a long-short term memory (LSTM) network], and wherein the executable instructions, when executed by the processor, further cause the processor to (Column 8 [lines 22-30] “FIG. 5 is a block diagram of the ensemble 300 of deep convolutional neural network pattern recognizers 306A, 306B, 306C, 306D and a module 310 which combines the outputs from each pattern recognizer to generate a final or ensemble score for each of the patches of FIG. 4. The ensemble of FIG. 5 can be considered a software system which may reside on a pathologist workstation or may alternatively reside on one or more computers in a local or wide area network”):
provide each of the patches to a corresponding one of the CNNs, and provide an output of each of the CNNs to the [LSTM network] (Column 9 [lines 13-14] “The outputs 308A-308D of each of the pattern recognizers 306A-306D is then combined in module 310.”),
[wherein the LSTM network is configured to extract the information representative of the hierarchical relationship that links the characteristics of the] histopathology image [present at the one level and at the another level] of the plurality of magnification levels (Column 8 line 53 continuing to column 9 line 3 “Four different networks (306A, 306B, 306C, 306C) were trained corresponding to 5×, 10×, 20×, and 40× magnification…Each network 306A, 306B, 306C, 306D generates its own output 308A, 308B, 308C, 308D in the form of a score between 0 and 1.”).
However, Stumpe fails to teach the LSTM network and wherein the LSTM network is configured to extract the information representative of the hierarchical relationship that links the characteristics of the [histopathology image] present at the one level and at the another level [of the plurality of magnification levels].
Zuo teaches the LSTM network and wherein the LSTM network is configured to extract the information representative of the hierarchical relationship that links the characteristics of the [histopathology image] present at the one level and at the another level [of the plurality of magnification levels] (Page 2988 right column
PNG
media_image1.png
52
314
media_image1.png
Greyscale
“where l∈[2,⋯,L] , and L is the number of scales. (rj,cj) is the position at the higher level scale j . h(rj,cj) is scale contextual element (already combined the four directional spatial dependencies, refer to Equation 14) from the higher level scale. Wjl is the scale j to scale l transformation matrix… b) HLSTM:
Similarly, for the HLSTM model (refer to Equation 15–18), the hidden representation of each gate functions is:
PNG
media_image2.png
130
328
media_image2.png
Greyscale
”)
Therefore, it would have been obvious to one of ordinary skill of the art before the effective filing date to modify Stumpe’s reference to include the LSTM network and wherein the LSTM network is configured to extract the information representative of the hierarchical relationship that links the characteristics of the [histopathology image] present at the one level and at the another level [of the plurality of magnification levels] taught by Zuo’s reference. The motivation for doing so would have been because dependencies among different image regions are very important for generating explicit image representations as suggested by Zuo (see Zuo, Abstract).
Further, one skilled in the art could have combined the elements described above by known methods with no change to the respective functions, and the combination would have yielded nothing more that predictable results. Therefore, it would have been obvious to combine Zuo with Stumpe to obtain the invention specified in claim 3.
Regarding claim 6, which claim 3 is incorporated, Stumpe discloses wherein the deep learning algorithm further comprises a Softmax activation function, and wherein the executable instructions, when executed by the processor, further cause the processor to (Column 9 [lines 9-11] “Multinomial logistical regression is known in the art of supervised learning and optimization, and is sometimes referred to as “Softmax Regression.”):
[provide an output of the LSTM network to the Softmax activation function], and generate a patch level classification based on an output of the Softmax activation function (Column 9 [lines 4-12] "The score between 0 and 1 is usually generated as the last layer of the neural network pattern recognizers 306A. 306B, 306C, 306D, in the form of a multinomial logistic regression, which generates a prediction, in the form of a probability of between 0 and 1, of which of the classes (here, healthy vs tumor) the input data (patch) belongs to. Multinomial logistical regression is known in the art of supervised learning and optimization, and is sometimes referred to as “Softmax Regression.”).
However, Stumpe fails to teach provide an output of the LSTM network to the Softmax activation function.
Zuo teaches provide an output of the LSTM network to the Softmax activation function (Page 2985 right column Section III first paragraph “consist of three types of layers: 1) five convolutional (and pooling) layers for extracting middle level image region features…Finally, an N-way (N indicates the number of categories) softmax loss layer is added on the top for classification”; Figure 1”).
Therefore, it would have been obvious to one of ordinary skill of the art before the effective filing date to modify Stumpe’s reference to include provide an output of the LSTM network to the Softmax activation function taught by Zuo’s reference. The motivation for doing so would have been because dependencies among different image regions are very important for generating explicit image representations as suggested by Zuo (see Zuo, Abstract).
Further, one skilled in the art could have combined the elements described above by known methods with no change to the respective functions, and the combination would have yielded nothing more that predictable results. Therefore, it would have been obvious to combine Zuo with Stumpe to obtain the invention specified in claim 6.
Regarding claim 7, which claim 6 is incorporated, Stumpe discloses wherein the Softmax activation function is configured to generate the output comprising a probability distribution over a set of predicted output classes (Column 9 [lines 4-12] "The score between 0 and 1 is usually generated as the last layer of the neural network pattern recognizers 306A. 306B, 306C, 306D, in the form of a multinomial logistic regression, which generates a prediction, in the form of a probability of between 0 and 1, of which of the classes (here, healthy vs tumor) the input data (patch) belongs to. Multinomial logistical regression is known in the art of supervised learning and optimization, and is sometimes referred to as “Softmax Regression.”).
Regarding claim 10, Stumpe discloses a non-transitory computer readable medium for detecting a medical condition in a histopathology image, the computer readable medium having program instructions for causing a hardware processor to (Column 15 [lines 48-51] "the present disclosure are directed to systems, apparatus, tangible, non-transitory computer-readable media, user interfaces, memory devices, and electronic devices for analyzing a tissue biopsy."):
obtain a plurality of patches at a plurality of magnification levels (patch at 40x, 20x, 10x, 5x in Column 6 [lines 45-57] equates to plurality of patches) from the histopathology image; apply a deep learning algorithm (four members in the ensemble in Column 6 [lines 45-57] equate to the deep learning algorithm) to each of the patches (Column 6 [lines 45-57] " each member of an ensemble of deep neural network pattern recognizers analyze pixel information of the patch, including some surrounding pixels, and generates a probability score of between 0 and 1. Each member operates on a patch of pixels but at different magnification levels. Preferably, there are four such members in the ensemble. For example, one operates on a 128×128 patch at 40× magnification. Another operates on a 128×128 patch but at 20× magnification (centered on or containing the 128×128 patch at 40×). A third one operates on a 128×128 patch but at 10× (again, centered on or containing the 128×128 patch at 40×). A fourth one operates on a 128×128 patch at 5×"; Figure 5);
[extract, from applying the deep learning algorithm, information representative of a hierarchical relationship that links characteristics of the] histopathology image [present at one level and another level] of the plurality of magnification levels (Column 8 line 53 continuing to column 9 line 3 “Four different networks (306A, 306B, 306C, 306C) were trained corresponding to 5×, 10×, 20×, and 40× magnification…Each network 306A, 306B, 306C, 306D generates its own output 308A, 308B, 308C, 308D in the form of a score between 0 and 1.”); and
identify the medical condition based on the extracted information [representative of the hierarchical relationship] for characteristics present at the one level and the another level of the plurality of magnification levels (Column 9 [lines 4-9] "The score between 0 and 1 is usually generated as the last layer of the neural network pattern recognizers 306A. 306B, 306C, 306D, in the form of a multinomial logistic regression, which generates a prediction, in the form of a probability of between 0 and 1, of which of the classes (here, healthy vs tumor) the input data (patch) belongs to.").
However, Stumpe fails to teach extract, from applying the deep learning algorithm, information representative of a hierarchical relationship that links characteristics of the [histopathology image] present at one level and another level [of the plurality of magnification levels].
Zuo teaches extract, from applying the deep learning algorithm (CNN in figure 1 equates to the deep learning algorithm), information representative of a hierarchical relationship that links characteristics of the [histopathology image] present at one level (lower level scales on Page 2988 equates to one level) and another level (the higher level scale Page 2988 equates to another level) [of the plurality of magnification levels] (Figure 1, Page 2988 last 2 paragraph of left column continuing to Right column first paragraph “local features are extracted from single-scale image regions. However, if the cross-scale information can be encoded, then better local descriptions can be achieved. Thus, we build recurrent connections across regions from different scales. For each element at each scale, its receptive field covers a number of elements at the lower level scales. More intuitively, as shown in the middle part of Figure 1, areas highlighted with yellow at the scale l+1 and l+2 are covered by the receptive field of the yellow element at the scale l . Thus, global information from the higher level scale l would be transferred to the corresponding areas at the lower level scales l+1 and l+2.”).
Therefore, it would have been obvious to one of ordinary skill of the art before the effective filing date to modify Stumpe’s reference to include extract, from applying the deep learning algorithm, information representative of a hierarchical relationship that links characteristics of the [histopathology image] present at one level and another level [of the plurality of magnification levels] taught by Zuo’s reference. The motivation for doing so would have been because dependencies among different image regions are very important for generating explicit image representations as suggested by Zuo (see Zuo, Abstract).
Further, one skilled in the art could have combined the elements described above by known methods with no change to the respective functions, and the combination would have yielded nothing more that predictable results. Therefore, it would have been obvious to combine Zuo with Stumpe to obtain the invention specified in claim 10.
Regarding claim 11 (drawn to a non-transitory computer readable medium), claim 11 is rejected the same as claim 2 and the arguments similar to that presented above for 2 are equally applicable to the claim 2, and all the other limitations similar to claim 2 are not repeated herein, but incorporated by reference.
Regarding claim 12 (drawn to a non-transitory computer readable medium), claim 12 is rejected the same as claim 3 and the arguments similar to that presented above for 3 are equally applicable to the claim 3, and all the other limitations similar to claim 3 are not repeated herein, but incorporated by reference.
Regarding claim 15 (drawn to a non-transitory computer readable medium), claim 15 is rejected the same as claim 6 and the arguments similar to that presented above for 6 are equally applicable to the claim 15, and all the other limitations similar to claim 6 are not repeated herein, but incorporated by reference.
Regarding claim 16 (drawn to a non-transitory computer readable medium), claim 16 is rejected the same as claim 7 and the arguments similar to that presented above for 7 are equally applicable to the claim 16, and all the other limitations similar to claim 6 are not repeated herein, but incorporated by reference.
Regarding claim 19, Stumpe discloses a method, comprising:
obtaining a plurality of patches at a plurality of magnification levels (patch at 40x, 20x, 10x, 5x in Column 6 [lines 45-57] equates to plurality of patches) from a histopathology image; applying a deep learning algorithm (four members in the ensemble in Column 6 [lines 45-57] equate to the deep learning algorithm) to each of the patches (Column 6 [lines 45-57] " each member of an ensemble of deep neural network pattern recognizers analyze pixel information of the patch, including some surrounding pixels, and generates a probability score of between 0 and 1. Each member operates on a patch of pixels but at different magnification levels. Preferably, there are four such members in the ensemble. For example, one operates on a 128×128 patch at 40× magnification. Another operates on a 128×128 patch but at 20× magnification (centered on or containing the 128×128 patch at 40×). A third one operates on a 128×128 patch but at 10× (again, centered on or containing the 128×128 patch at 40×). A fourth one operates on a 128×128 patch at 5×"; Figure 5);
[extracting, from applying the deep learning algorithm, information representative of a hierarchical relationship that links characteristics of the] histopathology image [present at one level and another level] of the plurality of magnification levels (Column 8 line 53 continuing to column 9 line 3 “Four different networks (306A, 306B, 306C, 306C) were trained corresponding to 5×, 10×, 20×, and 40× magnification…Each network 306A, 306B, 306C, 306D generates its own output 308A, 308B, 308C, 308D in the form of a score between 0 and 1.”); and
identifying a medical condition based on the extracted information [representative of the hierarchical relationship] for characteristics present at the one level and the another level of the plurality of magnification levels (Column 9 [lines 4-9] "The score between 0 and 1 is usually generated as the last layer of the neural network pattern recognizers 306A. 306B, 306C, 306D, in the form of a multinomial logistic regression, which generates a prediction, in the form of a probability of between 0 and 1, of which of the classes (here, healthy vs tumor) the input data (patch) belongs to.").
However, Stumpe fails to teach extracting, from applying the deep learning algorithm, information representative of a hierarchical relationship that links characteristics of the [histopathology image] present at one level and another level [of the plurality of magnification levels].
Zuo teaches extracting, from applying the deep learning algorithm (CNN in figure 1 equates to the deep learning algorithm), information representative of a hierarchical relationship that links characteristics of the [histopathology image] present at one level (lower level scales on Page 2988 equates to one level) and another level (the higher level scale Page 2988 equates to another level) [of the plurality of magnification levels] (Figure 1, Page 2988 last 2 paragraph of left column continuing to Right column first paragraph “local features are extracted from single-scale image regions. However, if the cross-scale information can be encoded, then better local descriptions can be achieved. Thus, we build recurrent connections across regions from different scales. For each element at each scale, its receptive field covers a number of elements at the lower level scales. More intuitively, as shown in the middle part of Figure 1, areas highlighted with yellow at the scale l+1 and l+2 are covered by the receptive field of the yellow element at the scale l . Thus, global information from the higher level scale l would be transferred to the corresponding areas at the lower level scales l+1 and l+2.”).
Therefore, it would have been obvious to one of ordinary skill of the art before the effective filing date to modify Stumpe’s reference to include extracting, from applying the deep learning algorithm, information representative of a hierarchical relationship that links characteristics of the [histopathology image] present at one level and another level [of the plurality of magnification levels] taught by Zuo’s reference. The motivation for doing so would have been because dependencies among different image regions are very important for generating explicit image representations as suggested by Zuo (see Zuo, Abstract).
Further, one skilled in the art could have combined the elements described above by known methods with no change to the respective functions, and the combination would have yielded nothing more that predictable results. Therefore, it would have been obvious to combine Zuo with Stumpe to obtain the invention specified in claim 19.
Regarding claim 20 (drawn to a method), claim 20 is rejected the same as claim 2 and the arguments similar to that presented above for 2 are equally applicable to the claim 2, and all the other limitations similar to claim 2 are not repeated herein, but incorporated by reference.
Claims 8 and 17 are rejected under 35 U.S.C. 103 as being unpatentable over Stumpe et al. (US 11,170,897 B2) (hereinafter, “Stumpe”) in view of Zuo et al. ("Learning contextual dependence with convolutional hierarchical recurrent neural networks." IEEE Transactions on Image Processing 25.7 (2016): 2983-2996.) (hereinafter, “Zuo”), and further in view of Pati et al. ("Hierarchical graph representations in digital pathology." Medical image analysis 75 (2022): 102264.) (hereinafter, “Pati”).
Regarding claim 8, which claim 1 is incorporated, Stumpe and Zuo both fail to teach wherein the hierarchical relationship that links characteristics of the histopathology image present at the plurality of magnification levels represents a relation between tissue morphology at a first one of the magnification levels and cell structure at a second one of the magnification levels.
Pati teaches wherein the hierarchical relationship that links characteristics of the histopathology image present at the plurality of magnification levels represents a relation between tissue morphology at a first one of the magnification levels and cell structure at a second one of the magnification levels (Page 5 right column Subsection 4.2.3 “Tissues in histopathology can be considered as hierarchical organizations of biological entities ranging from fine-level, i.e ., cells, to coarse-level, i.e ., tissue regions. There exist intra- and inter-level coupling based on topological distributions and interactions among the entities. Following this motivation, we propose HACT , a HierArchical Cell-to-Tissue ( HACT ) graph representation to jointly represent low-level CG and high-level TG…Inter-level topology is presented by a binary assignment (cell-to-tissue hierarchy) matrix…that utilizes the relative spatial distributions of nuclei with respect to tissue regions. For the i th nucleus and j th tissue region, the corresponding assignment is given as,
PNG
media_image3.png
76
720
media_image3.png
Greyscale
” Examiner interprets the hierarchal relationship as the binary assignment matrix which links nuclei centroids to tissue regions (i.e. cell structure to tissue morphology) ).
Therefore, it would have been obvious to one of ordinary skill of the art before the effective filing date to modify Stumpe in view of Zuo to include teaches wherein the hierarchical relationship that links characteristics of the histopathology image present at the plurality of magnification levels represents a relation between tissue morphology at a first one of the magnification levels and cell structure at a second one of the magnification levels taught by Pati’s reference. The motivation for doing so would have been to use both cell and tissue information to correctly identify the cancer subtype as suggested by Pati (see Pati, Page 4 left column first paragraph).
Further, one skilled in the art could have combined the elements described above by known methods with no change to the respective functions, and the combination would have yielded nothing more that predictable results. Therefore, it would have been obvious to combine Zuo and Pati with Stumpe to obtain the invention specified in claim 8
Regarding claim 17 (drawn to a non-transitory computer readable medium), claim 17 is rejected the same as claim 8 and the arguments similar to that presented above for 8 are equally applicable to the claim 8, and all the other limitations similar to claim 8 are not repeated herein, but incorporated by reference.
Claims 9 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Stumpe et al. (US 11,170,897 B2) (hereinafter, “Stumpe”) in view of Zuo et al. ("Learning contextual dependence with convolutional hierarchical recurrent neural networks." IEEE Transactions on Image Processing 25.7 (2016): 2983-2996.) (hereinafter, “Zuo”), and further in view of Ruan et al. ("A fast and effective detection framework for whole-slide histopathology image analysis." Plos one 16.5 (2021): e0251521.) (hereinafter, “Ruan”).
Regarding claim 9, which claim 1 is incorporated, Stumpe discloses [use an attention mechanism to identify a region of interest within the histopathology image, and] crop the histopathology image to generate the plurality of patches [based on the region of interest] (Column 5 [lines 34-48] "The slide image is subdivided into a multitude of rectangular patches, such as ˜12,000 or so. The number of patches per slide can range from 10,000 to 400,000. In one embodiment each patch is in the form of a square of 128×128 pixels or 299×299 pixels.").
However, Stumpe and Zuo both fail to teach use an attention mechanism to identify a region of interest within the histopathology image, [and crop the histopathology image to generate the plurality of patches] based on the region of interest.
Ruan teaches use an attention mechanism to identify a region of interest within the histopathology image (Page 10 second paragraph “The third case is that the generation algorithm focuses on searching for suspicious regions when fgrad is greater than Tgrad”), [and crop the histopathology image to generate the plurality of patches] based on the region of interest (Page 10 second paragraph “only the coordinates whose corresponding gradient is greater than fgrad will be selected. If the algorithm cannot find enough sample points that satisfy the constraint at once, then it will look again in the neighborhood of the sample points just selected.”).
Therefore, it would have been obvious to one of ordinary skill of the art before the effective filing date to modify Stumpe in view of Zuo to include teaches use an attention mechanism to identify a region of interest within the histopathology image, [and crop the histopathology image to generate the plurality of patches] based on the region of interest taught by Ruan’s reference. The motivation for doing so would have been to reduce gradient maps and enhance overall discovery capabilities as suggested by Ruan (see Ruan, Page 10 second paragraph).
Further, one skilled in the art could have combined the elements described above by known methods with no change to the respective functions, and the combination would have yielded nothing more that predictable results. Therefore, it would have been obvious to combine Zuo and Ruan with Stumpe to obtain the invention specified in claim 9.
Regarding claim 18, which claim 16 is incorporated, Stumpe discloses [use an attention mechanism to identify a region of interest within the histopathology image, and] crop the histopathology image to generate the plurality of patches [based on the region of interest] (Column 5 [lines 34-48] "The slide image is subdivided into a multitude of rectangular patches, such as ˜12,000 or so. The number of patches per slide can range from 10,000 to 400,000. In one embodiment each patch is in the form of a square of 128×128 pixels or 299×299 pixels.").
However, Stumpe and Zuo both fail to teach use an attention mechanism to identify a region of interest within the histopathology image, [and crop the histopathology image to generate the plurality of patches] based on the region of interest.
Ruan teaches use an attention mechanism to identify a region of interest within the histopathology image (Page 10 second paragraph “The third case is that the generation algorithm focuses on searching for suspicious regions when fgrad is greater than Tgrad”),
[and crop the histopathology image to generate the plurality of patches] based on the region of interest (Page 10 second paragraph “only the coordinates whose corresponding gradient is greater than fgrad will be selected. If the algorithm cannot find enough sample points that satisfy the constraint at once, then it will look again in the neighborhood of the sample points just selected.”).
Therefore, it would have been obvious to one of ordinary skill of the art before the effective filing date to modify Stumpe in view of Zuo to include teaches use an attention mechanism to identify a region of interest within the histopathology image, [and crop the histopathology image to generate the plurality of patches] based on the region of interest taught by Ruan’s reference. The motivation for doing so would have been to reduce gradient maps and enhance overall discovery capabilities as suggested by Ruan (see Ruan, Page 10 second paragraph).
Further, one skilled in the art could have combined the elements described above by known methods with no change to the respective functions, and the combination would have yielded nothing more that predictable results. Therefore, it would have been obvious to combine Zuo and Ruan with Stumpe to obtain the invention specified in claim 18.
Allowable Subject Matter
Claims 4 and 5 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.
Claims 13 and 14 would be allowable if rewritten to overcome the rejection(s) under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), 2nd paragraph, set forth in this Office action and to include all of the limitations of the base claim and any intervening claims.
Claims 4-5 and 13-14 contain subject matter that is not disclosed or made obvious in the cited art:
In regard to claim 4, when considering claim 4 as a whole, prior art fails to disclose or render obvious, alone or in combination:
“[…] sequentially learn, by providing, to the LSTM network, a result of the LSTM network processing an output of a preceding CNN each time the LSTM network is processing an output of a CNN other than an initial CNN, the hierarchical relationship that links the characteristics of the histopathology image present at the one level and at the another level of the plurality of magnification levels.”
In regard to claim 5, claim 5 depends on objected claim 4. Therefore, by virtue of their dependency, claim 5 is also indicated as objected subject matter.
In regard to claim 13, when considering claim 13 as a whole, prior art fails to disclose or render obvious, alone or in combination:
“[…] sequentially learn, by providing, to the LSTM network, a result of the LSTM network processing an output of a preceding CNN each time the LSTM network is processing an output of a CNN other than an initial CNN, the hierarchical relationship that links the characteristics of the histopathology image present at the one level and at the another level of the plurality of magnification levels.”
In regard to claim 14, claim 14 depends on claim 13. Therefore, by virtue of their dependency, claim 14 would be allowable if claim 13 is rewritten to overcome the rejection(s) under 35 U.S.C. 112(b).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Nahid et al. ("Histopathological breast cancer image classification by deep neural network techniques guided by local clustering." BioMed research international 2018.1 (2018): 2362108.) discloses using deep neural networks, including CNNs and LSTMs, to classify breast cancer histopathology images based on structural and statistical image features.
Fuchs et al. (US 2019/0295252 A1) discloses a system for generating tiles from a biomedical image and selecting the most relevant tile using an inference model based on a score indicating the likelihood of disease-related features.
Fuchs et al. (US 11,501,434 B2) discloses a Deep Multi-Magnification Networks that process image patches from multiple magnification levels using a multi-encoder and decoder architecture to improve tissue segmentation and support cancer diagnosis.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to UROOJ FATIMA whose telephone number is (571)272-2096. The examiner can normally be reached M-F 8:00-5:00.
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, Henok Shiferaw can be reached at (571) 272-4637. 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.
/UROOJ FATIMA/Examiner, Art Unit 2676
/CHINEYERE WILLS-BURNS/Supervisory Patent Examiner, Art Unit 2673
Prosecution Insights