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 .
This Office Action is responsive to the Reply to Office Action filed July 28, 2025. The Examiner acknowledges the amendments to claims 1 and 10, and the cancellation of claims 4-9, 13-20 and 23-24. Claims 1-3, 10-12, 21-22, and 25-26 are currently pending.
Response to Arguments
Applicant's arguments filed July 28, 2025 have been fully considered but they are not persuasive. Regarding Applicant’s argument that the rejection of the claims under 35 USC 101 should be withdrawn, the Examiner respectfully disagrees. Applicant alleges that the claims integrate the judicial exception into a practical application because the claims improve existing technologies for lung cancer diagnosis by improving the accuracy of early-stage lung cancer diagnosis and eliminating the need for invasive procedures and repeated radiation exposure associated with current lung cancer diagnostic methods. Applicant has not provided sufficient evidence of an improvement, and the mere assertion that there is an improvement to the technological field is not persuasive. Moreover, the additional elements of the invention use well-understood, routine, and conventional (WURC) functions that do not contribute as an improvement to the technology field. Further, the additional elements described in the claims, either alone, or in combination, fail to integrate the judicial exception into a practical application. Therefore, the rejection of the claims under 35 USC 101 still stands. See 35 USC 101 section below.
Regarding Applicant’s arguments against the previous rejection of the claims under 35 USC 103, Applicant argues that El-Baz fails to teach a quantified appearance of the anatomical structure, and instead teaches using only spherical harmonic shape analysis of an anatomical structure for diagnosing lung nodules. Further, Applicant argues that El-Baz fails to teach spherical harmonic shape analysis of an anatomical structure and a quantified vector appearance of the anatomical structure as separate indicators. Moreover, Applicant argues that the combination of Cohen, El-Baz and Fu fails to teach generating a plurality of initial classification probabilities from each of the plurality of image-based measurable indicators because El-Baz fails to teach generating initial classification probabilities from each of a spherical harmonic shape analysis and a quantified appearance of the anatomical structure.
The Examiner respectfully disagrees with Applicant’s argument that the cited references fail to teach generating a plurality of initial classification probabilities from each of the plurality of image-based measurable indicators, and further that El-Baz fails to teach a quantified appearance of the anatomical structure, and instead teaches using only spherical harmonic shape analysis of an anatomical structure for diagnosing lung nodules. Cohen teaches generating an initial score for a subject regarding the presence/absence of cancer based on patient data using a neural network (Cohen, see par 0112). The initial score is the initial classification as to whether or not a patient has the presence or absence of cancer. Moreover, Cohen teaches that one or more neural networks may be used to classify an individual patient into one of a plurality of categories (i.e., a likelihood of cancer), wherein the neural networks use biomarkers and clinical parameters such as imaging data, the size of lung nodules and the number of nodules (see Cohen, par 0202-0208 & 0268). Moreover, El-Baz does teach the use of appearance models, specifically two probabilistic visual appearance models that are a learned lung nodule appearance prior and a current appearance model of the image to be segmented (see El-Baz, Col. 7, lines 35-56, Col. 8, lines 9-21, Col. 9, lines 45-67, Col. 14, lines 11-25). However, the Examiner does agree that the cited references fail to teach generating initial classification probabilities from each of a spherical harmonic shape analysis and a quantified appearance of the anatomical structure. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of newly found prior art that in combination with the other prior art references, teaches the amended claim limitations of generating initial classification probabilities from . See 35 USC 103 rejections below.
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 1-3, 10-12, 21-22 & 25-26 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Claim(s) 1 & 10 recite(s), at least in part the following step(s): “generating a plurality of initial classification probabilities, one from each of the plurality of image-based measurable indicators and from the biomarker-based measurable indicator”, “assigning a weight to each initial classification probability”, and “generating a final classification probability by integrating the plurality of initial classification probabilities based on their respective weights”. These step(s), when given its/their broadest reasonable interpretation(s), describe(s) carrying out said step(s) mentally (i.e. a mental task in the human mind, and/or by a mathematical process) but for the recitation of generic computer components. In other words, absent the recitation of a neural network, nothing precludes the claimed step from practically being performed mentally (i.e. a mental task in the human mind, and/or by a mathematical process). For example, absent the limitation(s) of a neural network used to generate a plurality of initial classification probabilities, to assign weights to each initial classification probability, and to generate a final classification probability in the step(s), the “generating” and “assigning” in the step(s) involves the user manually using pen and paper, mentally, visually and/or by a mathematical process, function, or equation such as the attraction-repulsion algorithm, Markov-Gibbs random field, or the classification probability equation as discussed in the specification of the instant application. In view of the foregoing, claim(s) 1-3, 10-12, 21-22 & 25-26 recite(s) an abstract idea.
For example, the following caselaw: Elec. Power Grp., LLC v. Alstom S.A. (Fed. Cir. 2016) contains the following analysis: “Information as such is an intangible. See Microsoft Corp. v. AT & T Corp., 550 U.S. 437, 451 n.12 (2007). Accordingly, we have treated collecting information, including when limited to particular content (which does not change its character as information), as within the realm of abstract ideas. See, e.g., Internet Patents, 790 F.3d at 1349; OIP Techs., Inc. v. Amazon. com, Inc., 788 F.3d 1359, 1363 (Fed. Cir. 2015). In a similar vein, we have treated analyzing information by steps people go through in their minds, or by mathematical algorithms, without more, as essentially mental processes within the abstract-idea category. See, e.g., TLI Commc’ns, 823 F.3d at 613; Digitech, 758 F.3d at 1351; SmartGene, Inc. v. Advanced Biological Labs., SA, 555 F. App’x 950, 955 (Fed. Cir. 2014); Bancorp Servs., L.L.C. v. Sun Life Assurance Co. of Canada (U.S.), 687 F.3d 1266, 1278 (Fed. Cir. 2012); CyberSource Corp. v. Retail Decisions, Inc., 654 F.3d 1366, 1372 (Fed. Cir. 2011); SiRF Tech., Inc. v. Int’l Trade Comm’n, 601 F.3d 1319, 1333 (Fed. Cir. 2010); see also Mayo, 132 S. Ct. at 1301; Parker v. Flook, 437 U.S. 584, 589–90 (1978); Gottschalk v. Benson, 409 U.S. 63, 67 (1972); Diamond v. Diehr, 450 U.S. 175 (1981). And we have recognized that merely presenting the results of abstract processes of collecting and analyzing information, without more (such as identifying a particular tool for presentation), is abstract as an ancillary part of such collection and analysis. See, e.g., Content Extraction, 776 F.3d at 1347; Ultramercial, Inc. v. Hulu, LLC, 772 F.3d 709, 715 (Fed. Cir. 2014). Here, the claims are clearly focused on the combination of those abstract-idea processes. The advance they purport to make is a process of gathering and analyzing information of a specified content, then displaying the results, and not any particular assertedly inventive technology for performing those functions. They are therefore directed to an abstract idea.” [Emphasis added].
The judicial exception(s) is/are not integrated into a practical application. Particularly, the claim(s) recite(s) the following additional element(s): ”extracting a plurality of measurable indicators of the presence or absence of a lung cancer disease state in a subject from the image data, wherein the plurality of image-based measurable indicators include three separate image-based measurable indicators: spherical harmonic shape analysis of an anatomical structure of the subject, a size of the anatomical structure, and a quantified appearance of the anatomical structure” and “receiving a biomarker-based measurable indicator of the presence or absence of a lung cancer disease state in a subject, wherein the biomarker-based indicator is one or more subject values of volatile organic compounds in an exhaled breath sample from the subject”. Neither the arrangement of the additional elements, nor the additional elements themselves, apply, rely on, or use the judicial exception recited supra in a manner that imposes a meaningful limit on the judicial exception. Rather, the additional element(s) is/are recited with a high level of generality (i.e., as a generic pre-solution data collection means performing a generic data collection function) such that it amounts to no more than instructions to apply the exception using a generic computer component. Therefore, the additional element(s) do(es) not integrate the exception(s) into a practical application because it does not impose any meaningful limits on practicing the abstract idea. The claim is directed to an abstract idea and/or law of nature.
The claim(s) include(s) the additional step(s)/element(s) recited above. The additional step(s)/element(s) are not sufficient to amount to significantly more than the judicial exception(s) since such additional step(s)/element(s) are generically claimed to enable an insignificant extra-solution activity including the collection of data by performing the basic functions of: (i) receiving, processing, and/or calculating data, and/or (ii) automating mental tasks. For example, in paragraphs 0027 and 0038 of the instant application, the applicant admits that breath samples collected from patients are obtained using a non-reactive TedlarTM bag from Sigma Aldrich, St Louis, Mo. The courts have recognized these functions to be well-understood, routine, and conventional functions when claimed in a merely generic manner. Therefore, the Office takes Official notice that the instantly claimed additional steps/elements are well-understood, routine and convention. Merely adding hardware that performs ‘“well understood, routine, conventional activities]’ previously known to the industry” will not make claims patent-eligible (In re TLI Communications LLC). In other words, the additional step(s)/element(s) amount(s) to no more than mere instructions to apply the exception(s) using generic computer component(s). Mere instructions to apply an exception using a generic computer component cannot provide an inventive concept. Accordingly, Claim(s) 1-18 do(es) not amount to significantly more than the abstract idea itself.
In regards to claim(s) 2-3, 11-12, 21-22 & 25-26, the claimed invention further describes the judicial exception in detail without however integrating said judicial exception into a practical application and/or providing additional elements that are sufficient to amount to significantly more than the judicial exception for reasons provided supra.
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.
Claim(s) 1, 10, 21-22, & 25-26 is/are rejected under 35 U.S.C. 103 as being unpatentable over US Patent Application Publication 20180068083 -- as previously cited--, hereinafter referenced as "Cohen" in view of US Patent 9230320 --as previously cited--, hereinafter referenced as "El-Baz" and further in view of US Patent Application Publication 20100111386, hereinafter referenced as “El-Baz ‘386” and US Patent 9638695 --as previously recited--, hereinafter referenced as "Fu".
With respect to claim 1, Cohen teaches a computer-aided method for identifying the presence or absence of a cancer disease state (see Cohen, par 0017-0018), the method comprising:
receiving a plurality of measurable indicators (i.e., biomarkers, clinical parameters, collected blood samples, or other types of patient records) of the presence or absence of a cancer disease state in a subject (see Cohen, par 0018, 0021, 0027-0029);
generating a plurality of initial classification probabilities (i.e., generating an initial score for a subject regarding the presence/absence of cancer based on patient data), using neural networks (i.e., a machine learning system), from each of the plurality of measurable indicators, wherein at least one initial classification probability is generated by a first neural network specific to that indicator (Cohen, see par 0112 & 0202-0208);
assigning a weight to each initial classification probability (i.e., quantifying the increased risk for the presence of cancer for a subject based upon weighting of risk factors that increase the likelihood of having cancer) using the neural network (i.e., machine learning system) (Cohen, see par 0112);
and generating a final classification by integrating the plurality of initial classification probabilities based on their respective weights using the neural network (i.e., providing a risk score for the subject, wherein the quantified increased risk for the presence of cancer is used to provide the risk score) (see Cohen, par 0112);
wherein the final classification is designating the presence or absence of a cancer disease state in the subject (i.e., providing a risk score for the presence of cancer) (Cohen, see par 0112).
Cohen further teaches the plurality of measurable indicators includes a size of the anatomical structure (i.e., size of pulmonary nodules) and that one or more neural networks may be used to classify an individual patient into one of a plurality of categories (i.e., a likelihood of cancer), wherein the neural networks use biomarkers and clinical parameters such as imaging data (see Cohen, fig. 18-19, par 0024, 0202-0208, 0443, 0446, 0465, 0471-0472).
Cohen fails to teach that the received image data is image data of a lung of a subject having at least one lung nodule, that the lung nodule is segmented from the image data, that image-based measurable indicators are extracted from the image data, that the initial classification probability is based on biomarker-based measurable indicators, wherein the initial classification probability is generated by a first neural network, nor that weights are assigned to each initial classification probability using a second neural network, nor that the final classification using weights from the second neural network and designates the presence or absence of a lung cancer disease state. Cohen further fails to teach that the plurality of image-based measurable indicators include three separate image-based measurable indicators (i.e., a spherical harmonic shape analysis of an anatomical structure of the subject and a quantified appearance of the anatomical structure, in addition to the size of the anatomical structure), and that the biomarker-based indicator is one or more subject values of volatile organic compounds in an exhaled breath sample from a subject.
El-Baz teaches a method for the detection of lung cancer, wherein a plurality of measurable indicators of the presence or absence of a lung cancer disease state include image-based biomarkers derived from CT scans, such as the appearance and shape of lung nodules, and clinical-based biomarker data that is used to train the neural network (see El-Baz, Col. 4, lines 11-40, Col. 7, lines 49-67, Col. 8, lines 1-47, Col. 13, lines 41-59). Furthermore, lung nodules are segmented from the image data (see El-Baz, Col. 5, lines 1-20). Moreover, a plurality of neural networks are used such as an appearance diagnostic network, a shape diagnostic network, an initial classification network and final diagnosis network that assigns a weight to each initial classification and generates a final classification by integrating the classification probabilities based upon their respective weights, using the image-based biomarker data and the clinical-based biomarker data (i.e., data learned from training) (see El-Baz, Col. 2, lines 5-38, Col. 7, lines 35-67, Col. 8, lines 1-67, Col. 9, lines , 1-67, Col. 10, lines 1-34). Further, El-Baz teaches that spherical harmonic shape analysis of an anatomical structure of the subject (i.e., spherical harmonic shape analysis of pulmonary nodules of a subject) (see El-Baz, fig. 2G-2I, Col. 5, lines 21-51) using a Markov-Gibbs random field model of the intensities of the lung nodules of the anatomical structure (see El-Baz, figs. 8A-8L, Col. 7, lines 35-48, Col. 7, lines 49-56) is used as a measurable indicator in a computer-aided diagnostic system for diagnosing malignant lung nodules using neural networks.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Cohen such that received image data is image data of a lung of a subject having at least one lung nodule, that the lung nodule is segmented from the image data, that image-based measurable indicators are extracted from the image data, that the initial classification probability is based on biomarker-based measurable indicators, wherein the initial classification probability is generated by a first neural network, that weights are assigned to each initial classification probability using a second neural network, and that the final classification uses weights from the second neural network and designates the presence or absence of a lung cancer disease state because that integrates both image-based and clinical biomarker data using a plurality of neural networks that aid in providing the most accurate classification of whether or not a lung nodule is benign or malignant, and therefore whether or not a patient has a presence of a lung cancer disease state (see El-Baz, Col. 4, lines 11-40, Col. 7, lines 49-67, Col. 8, lines 1-47, Col. 13, lines 41-59).
Additionally, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Cohen such that spherical harmonic shape analysis of an anatomical structure of the subject is used because the implementation of spherical harmonic shape analysis enables the computer-aided diagnostic system to accurately predict the presence or absence of cancer based upon data representing the shape and appearance of the anatomical structures, in addition to the size of the anatomical structure (see El-Baz, fig. 2G-2I, Col. 5, lines 21-51, figs. 8A-8L, Col. 7, lines 35-48, Col. 7, lines 49-56).
Cohen in view of El-Baz fails to teach that the plurality of image-based measurable indicators include three separate image-based measurable indicators (i.e., a quantified appearance of the anatomical structure, in addition to the size of the anatomical structure and spherical harmonic shape analysis of the anatomical structure), and that the biomarker-based indicator is one or more subject values of volatile organic compounds in an exhaled breath sample from the subject.
El-Baz ‘386 teaches a computer aided diagnostic system incorporating lung segmentation and registration wherein the visual appearance of a lung in low-dose spiral computed tomography (LDCT) images is modeled by a Markov-Gibbs random field (MGRF) (see El-Baz ‘386, par 0105-0108, fig. 8), and this quantified visual appearance of the lung is used to register the lung so that the computer aided diagnostic system can calculate volumetric changes in detected lung nodules and accurately calculate the rate of growth of pulmonary nodules (see El-Baz ‘386, par 0105-0108).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Cohen as modified by El-Baz such that that the plurality of image-based measurable indicators include three separate image-based measurable indicators (i.e., a quantified appearance of the anatomical structure, in addition to the size of the anatomical structure and spherical harmonic shape analysis of the anatomical structure) because a quantified visual appearance of a lung can be used to register the lung in a computer aided diagnostic system, which permits the calculation of volumetric changes in detected lung nodules and the calculation of the rate of growth of lung nodules (see El-Baz ‘386, par 0105-0108).
Cohen in view of El-Baz and El-Baz ‘386 fails to teach the biomarker-based indicator is one or more subject values of volatile organic compounds in an exhaled breath sample from the subject.
Fu teaches that subject values of volatile organic compounds in an exhaled breath sample from a subject is a non-invasive method of detecting or screening for a lung cancer disease state in a subject (see Fu, Col. 2, lines 33-53, Col. 5, lines 1-22).
It would have been obvious to one of ordinary skill in the art before the effective filing date of
the claimed invention to use the known technique of subject values of volatile organic compounds in an
exhaled breath sample, as taught by Fu, in the computer-implemented diagnostic method for cancer, as
taught by Cohen in view of El-Baz and El-Baz ‘386, that includes measurable indicators of size, spherical harmonic analysis, a quantified appearance of an anatomical structure of the subject, and subject values of volatile organic compounds in an exhaled breath sample because the additional measurable indicator of volatile organic compounds in an exhaled breath samples improves the predictive diagnostic capability of the computer-implemented diagnostic method by providing another metric by which the neural network employs to predict the presence or absence of cancer (see Fu, Col. 2, lines 33-53, Col. 5, lines 1-22).
With respect to claim 10, Cohen teaches a non-transitory (i.e., non-volatile memory) computer readable storage medium (Cohen, see par 0017-0018 & 0196) having computer program instructions stored thereon that, when executed by a processor, cause the processor (see Cohen, par 0017-0018 & 0196) to perform the following instructions:
receiving a plurality of measurable indicators (i.e., biomarkers, clinical parameters, collected blood samples, or other types of patient records) of the presence or absence of a cancer disease state in a subject (see Cohen, par 0018, 0021, 0027-0029, 0202-0208);
generating a plurality of initial classification probabilities (i.e., generating an initial score for a subject regarding the presence/absence of cancer based on patient data), using neural networks (i.e., a machine learning system), from each of the plurality of measurable indicators (Cohen, see par 0112);
assigning a weight to each initial classification probability (i.e., quantifying the increased risk for the presence of cancer for a subject based upon weighting of risk factors that increase the likelihood of having cancer) using the neural network (i.e., machine learning system) (Cohen, see par 0112);
and generating a final classification by integrating the plurality of initial classification probabilities based on their respective weights using the neural network (i.e., providing a risk score for the subject, wherein the quantified increased risk for the presence of cancer is used to provide the risk score) (see Cohen, par 0112);
wherein the final classification is designating the presence or absence of a cancer disease state in the subject (i.e., providing a risk score for the presence of cancer) (Cohen, see par 0112).
Cohen further teaches the plurality of measurable indicators includes a size of the anatomical structure (i.e., size of pulmonary nodules) and that one or more neural networks may be used to classify an individual patient into one of a plurality of categories (i.e., a likelihood of cancer), wherein the neural networks use biomarkers and clinical parameters such as imaging data (see Cohen, fig. 18-19, par 0024, 0202-0208, 0443, 0446, 0465, 0471-0472).
Cohen fails to teach that the received image data is image data of a lung of a subject having at least one lung nodule, that the lung nodule is segmented from the image data, that image-based measurable indicators are extracted from the image data, that the initial classification probability is based on biomarker-based measurable indicators, wherein the initial classification probability is generated by a first neural network, nor that weights are assigned to each initial classification probability using a second neural network, nor that the final classification using weights from the second neural network and designates the presence or absence of a lung cancer disease state. Cohen further fails to teach that the plurality of image-based measurable indicators include three separate image-based indicators (i.e., a spherical harmonic shape analysis of an anatomical structure of the subject and a quantified appearance of the anatomical structure, in addition to the size of the anatomical structure of the subject), and that the biomarker-based indicator is one or more subject values of volatile organic compounds in an exhaled breath sample from a subject.
El-Baz teaches a method for the detection of lung cancer, wherein a plurality of measurable indicators of the presence or absence of a lung cancer disease state include image-based biomarkers derived from CT scans, such as the appearance and shape of lung nodules, and clinical-based biomarker data that is used to train the neural network (see El-Baz, Col. 4, lines 11-40, Col. 7, lines 49-67, Col. 8, lines 1-47, Col. 13, lines 41-59). Furthermore, lung nodules are segmented from the image data (see El-Baz, Col. 5, lines 1-20). Moreover, a plurality of neural networks are used such as an appearance diagnostic network, a shape diagnostic network, an initial classification network and final diagnosis network that assigns a weight to each initial classification and generates a final classification by integrating the classification probabilities based upon their respective weights, using the image-based biomarker data and the clinical-based biomarker data (i.e., data learned from training) (see El-Baz, Col. 2, lines 5-38, Col. 7, lines 35-67, Col. 8, lines 1-67, Col. 9, lines , 1-67, Col. 10, lines 1-34). Further, El-Baz teaches that spherical harmonic shape analysis of an anatomical structure of the subject (i.e., spherical harmonic shape analysis of pulmonary nodules of a subject) (see El-Baz, fig. 2G-2I, Col. 5, lines 21-51) using a Markov-Gibbs random field model of the intensities of the lung nodules of the anatomical structure (see El-Baz, figs. 8A-8L, Col. 7, lines 35-48, Col. 7, lines 49-56) is used as a measurable indicator in a computer-aided diagnostic system for diagnosing malignant lung nodules using neural networks.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Cohen such that received image data is image data of a lung of a subject having at least one lung nodule, that the lung nodule is segmented from the image data, that image-based measurable indicators are extracted from the image data, that the initial classification probability is based on biomarker-based measurable indicators, wherein the initial classification probability is generated by a first neural network, that weights are assigned to each initial classification probability using a second neural network, and that the final classification uses weights from the second neural network and designates the presence or absence of a lung cancer disease state, because that would improve the system of Cohen by integrating both image-based and clinical biomarker data using a plurality of neural networks that aid in providing the most accurate classification of whether or not a lung nodule is benign or malignant, and therefore whether or not a patient has a presence of a lung cancer disease state (see El-Baz, Col. 4, lines 11-40, Col. 7, lines 49-67, Col. 8, lines 1-47, Col. 13, lines 41-59).
Additionally, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Cohen such that spherical harmonic shape analysis of an anatomical structure of the subject is used because the implementation of spherical harmonic shape analysis enables the computer-aided diagnostic system to accurately predict the presence or absence of cancer based upon data representing the shape and appearance of the anatomical structures, in addition to the size of the anatomical structure (see El-Baz, fig. 2G-2I, Col. 5, lines 21-51, figs. 8A-8L, Col. 7, lines 35-48, Col. 7, lines 49-56).
Cohen in view of El-Baz fails to teach that the plurality of image-based measurable indicators include three separate image-based measurable indicators (i.e., a quantified appearance of the anatomical structure, in addition to the size of the anatomical structure of the subject and spherical harmonic shape analysis of the anatomical structure of the subject), and that the biomarker-based indicator is one or more subject values of volatile organic compounds in an exhaled breath sample from the subject.
El-Baz ‘386 teaches a computer aided diagnostic system incorporating lung segmentation and registration wherein the visual appearance of a lung in low-dose spiral computed tomography (LDCT) images is modeled by a Markov-Gibbs random field (MGRF) (see El-Baz ‘386, par 0105-0108, fig. 8), and this quantified visual appearance of the lung is used to register the lung so that the computer aided diagnostic system can calculate volumetric changes in detected lung nodules and accurately calculate the rate of growth of pulmonary nodules (see El-Baz ‘386, par 0105-0108).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Cohen as modified by El-Baz such that that the plurality of image-based measurable indicators include three separate image-based measurable indicators (i.e., a quantified appearance of the anatomical structure, in addition to the size of the anatomical structure and spherical harmonic shape analysis of the anatomical structure) because a quantified visual appearance of a lung can be used to register the lung in a computer aided diagnostic system, which permits the calculation of volumetric changes in detected lung nodules and the calculation of the rate of growth of lung nodules (see El-Baz ‘386, par 0105-0108).
Cohen in view of El-Baz and El-Baz ‘386 fails to teach the biomarker-based indicator is one or more subject values of volatile organic compounds in an exhaled breath sample from the subject.
Fu teaches that subject values of volatile organic compounds in an exhaled breath sample from a subject is a non-invasive method of detecting or screening for a lung cancer disease state in a subject (see Fu, Col. 2, lines 33-53, Col. 5, lines 1-22).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to use the known technique of subject values of volatile organic compounds in an exhaled breath sample, as taught by Fu, in the computer-implemented diagnostic method for cancer, as taught by Cohen in view of El-Baz and El-Baz ‘386, that includes measurable indicators of size, spherical harmonic analysis, a quantified appearance of an anatomical structure of the subject, and subject values of volatile organic compounds in an exhaled breath sample because the additional measurable indicator of volatile organic compounds in an exhaled breath samples improves the predictive diagnostic capability of the computer-implemented diagnostic method by providing another metric by which the neural network employs to predict the presence or absence of cancer (see Fu, Col. 2, lines 33-53, Col. 5, lines 1-22).
With respect to claim 21, Cohen as modified by El-Baz, El-Baz ‘386 and Fu teaches the method of claim 1. Cohen as modified by El-Baz, El-Baz ‘386 and Fu further teaches the anatomical structure is the at least one lung nodule (see El-Baz, Col. 4, lines 11-40).
With respect to claim 22, Cohen as modified by El-Baz, El-Baz ‘386 and Fu teaches the method of claim 1. Cohen as modified by El-Baz, El-Baz ‘386 and Fu further teaches the spherical harmonic shape analysis comprises determining the number of spherical harmonics required to approximate the structure of the anatomical structure (see El-Baz, Col. 4, lines 11-40).
With respect to claim 25, Cohen as modified by El-Baz, El-Baz ‘386 and Fu teaches the non-transitory computer readable storage medium of claim 10. Cohen as modified by El-Baz, El-Baz ‘386 and Fu further teaches the anatomical structure is the at least one lung nodule (see El-Baz, Col. 4, lines 11-40).
With respect to claim 26, Cohen as modified by El-Baz, El-Baz ‘386 and Fu teaches the non-transitory computer readable storage medium of claim 10. Cohen as modified by El-Baz, El-Baz ‘386 and Fu further teaches the spherical harmonic shape analysis comprises determining the number of spherical harmonics required to approximate the structure of the anatomical structure (see El-Baz, Col. 4, lines 11-40).
Claim(s) 2-3 & 11-12 is/are rejected under 35 U.S.C. 103 as being unpatentable over Cohen in view of El-Baz, El-Baz '386 and Fu and in further view of US Patent Application Publication 20180144465 --as previously recited--, hereinafter referenced as "Hsieh".
With respect to claim 2 Cohen as modified by El-Baz, El-Baz ‘386 and Fu teaches the method of claim 1.
Cohen as modified by El-Baz, El-Baz ‘386 and Fu fails to teach that each of the first and second neural networks includes a dimensionality reducer and a softmax layer.
Hsieh teaches that neural networks used for medical applications, such as the prediction of a cancer disease state (i.e., lung cancer) (see Hsieh par 0177, 0295) includes a dimensionality reducer (i.e., an autoencoder network) (see Hsieh, par 0137, 0194) and a softmax layer (see Hsieh, par 0230, 0233-0234).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Cohen as modified by El-Baz, El-Baz ‘386 and Fu such that each of the first and second neural networks includes a dimensionality reducer and a softmax layer because a dimensionality reducer and softmax layer in each stage of the neural network allows for new data inputted into the network, such as images, to be leveraged against prior data acquisitions in order to generate a predictive output (i.e., a probability distribution) using the softmax classifier layer (see Hsieh, par 0137, 0194, 0230, 0233-0234), thus improving the accuracy of the predictive output of the neural network.
With respect to claim 3, Cohen as modified by El-Baz, El-Baz ‘386, Fu and in further view of Hsieh teaches the method of claim 2. Cohen as modified by El-Baz, El-Baz ‘386 and Fu fails to teach the dimensionality reducer used in the neural network is an autoencoder.
Hsieh teaches that autoencoders are used as a dimensionality reducer in neural networks used for medical applications, such as the prediction of a cancer disease state (i.e., lung cancer) (see Hsieh, par 0177, 0137, 0194, & 0295).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Cohen as modified by El-Baz, El-Baz ‘386 and Fu such that the dimensionality reducer is an autoencoder because an autoencoder enables network layers to be initialized even when prior knowledge (i.e., data) does not exist (see Hsieh, par 0135), and enables representations or encodings to be learned for a set of data (see Hsieh, par 0137).
With respect to claim 11, Cohen as modified by El-Baz, El-Baz ‘386 and Fu teaches the non-transitory computer readable storage medium (i.e., non-volatile memory) of claim 10. Cohen as modified by El-Baz, El-Baz ‘386 and Fu fails to teach that each of the first and second neural networks includes a dimensionality reducer and a softmax layer.
Hsieh teaches that neural networks used for medical applications, such as the prediction of a cancer disease state (i.e., lung cancer) (see Hsieh par 0177, 0295) includes a dimensionality reducer (i.e., an autoencoder network) (see Hsieh, par 0137, 0194) and a softmax layer (see Hsieh, par 0230, 0233-0234).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Cohen as modified by El-Baz, El-Baz ‘386 and Fu such that each of the first and second neural networks includes a dimensionality reducer and a softmax layer because a dimensionality reducer and softmax layer in each stage of the neural network allows for new data inputted into the network, such as images, to be leveraged against prior data acquisitions in order to generate a predictive output (i.e., a probability distribution) using the softmax classifier layer (see Hsieh, par 0137, 0194, 0230, 0233-0234), thus improving the accuracy of the predictive output of the neural network.
With respect to claim 12, Cohen as modified by El-Baz, El-Baz ‘386 and Fu and in further view of Hsieh teaches the non-transitory computer readable storage medium of claim 11. Cohen as modified by El-Baz, El-Baz ‘386 and Fu fails to teach that the dimensionality reducer used in the neural network is an autoencoder.
Hsieh teaches that autoencoders are used as a dimensionality reducer in neural networks used for medical applications, such as the prediction of a cancer disease state (i.e., lung cancer) (see Hsieh, par 0177, 0137, 0194, & 0295).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the method of Cohen as modified by El-Baz, El-Baz ‘386 and Fu such that the dimensionality reducer is an autoencoder because an autoencoder enables network layers to be initialized even when prior knowledge (i.e., data) does not exist (see Hsieh, par 0135), and enables representations or encodings to be learned for a set of data (see Hsieh, par 0137).
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 Destiny J Cruickshank whose telephone number is (571)270-0187. The examiner can normally be reached M-F, 9am-6pm.
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, Charles Marmor II can be reached at (571) 272-4730. 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.
/CHARLES A MARMOR II/Supervisory Patent Examiner
Art Unit 3791
/D.J.C./Examiner, Art Unit 3791