DETAILED ACTION
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 1-20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception without significantly more. The claim(s) recite(s) limitations that fall under the grouping of abstract idea of “Certain Methods of Organizing Human Activity”, e.g. Concepts Relating To Managing Human Behavior (Step 2A, Prong One) and “Mental Processes”, e.g. concepts performed in the human mind (including an observation, evaluation, judgment, opinion) (step 2A). Specifically, the claim recites various steps of human activity that can be performed in the mind or with pen and paper, such as, recites receiving an image, determining a diagnosis, determining whether the diagnosis is positive for a disease, determining a subclassification of the diagnosis, and outputting a result based on the subclassification. These operations amount to obtaining, evaluating, and classifying information.
Under step 2A, prong two, this judicial exception is not integrated into a practical application. The claim does not recite any specific computing implementation, does not require any particular algorithm, model or hardware configuration, and lacks technical improvement to image processing or machine learning. Furthermore, the judicial exception is not integrated into a practical application because the claims are directed to an abstract idea with additional generic computer elements (e.g. processor, memory, computer storage medium, etc.), which are generically recited computer elements that do not add a meaningful limitation to the abstract idea because they amount to simply implementing the abstract idea on a computer. The claims merely apply the evaluation and classification process to medical images using diagnostic models, subclassification models, extraction of biomarkers, and a rules engine.
Under step 2B, the claims does/do not include additional elements that are sufficient to amount to significantly more than the judicial exception because these are well-understood, routine, conventional computer functions as recognized by the court decisions listed in MPEP § 2106.05(d). The claims do not include an inventive concept that is sufficient to transform the abstract idea into a patent-eligible application. The claimed diagnostic models, subclassification models, biomarker extraction, rules engine, computer-readable medium, memory, and processors merely implement the identified classification process using generic computer components and do not provide a technological improvement.
The additional limitations of claims 2-7, 9-14, and 16-20 likewise merely specify particular information, models, training data, rules, or outputs used in performing the same classification process and do not integrate the abstract idea into a practical application.
Claims 8-14 (drawn to a CRM) are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter.
Claims 8-14 covers both statutory and non-statutory embodiments (under the broadest reasonable interpretation of the claim when read in light of the specification and in view of one skilled in the art) and embraces subject matter that is not eligible for patent protection and therefore is directed to non-statutory subject matter.
Specifically, the Specification at ¶34-36 wherein “Such a computer program may be stored in a non-transitory, tangible computer readable storage medium, or any type of media suitable for storing electronic instructions, which may be coupled to a computer system bus” and “Such a product may comprise information resulting from a computing process, where the information is stored on a non-transitory, tangible computer readable storage medium and may include any embodiment of a computer program product or other data combination described herein” given the broadest reasonable interpretation does not exclude a signal. Thus, the claims are not eligible subject matter. It is recommended to amend and narrow the claims to cover only statutory embodiments to avoid a rejection under 35 U.S.C. § 101 by adding the limitation "non- transitory" to the claims.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 1, 5, 7, 8, 12, 14, 15 and 19 is/are rejected under 35 U.S.C. 103 as being unpatentable over An et al (NPL: Hierarchical deep learning models using transfer learning for disease detection and classification based on small number of medical images) in view of Mavroeidis et al (US 20200175677).
Regarding claim 1, An discloses a method for sub-classifying a diagnosis (Fig. 2b Proposed approach 2: hierarchical classification is used to create a low-level model (Model 2) for classifying normal versus disease cases and a high-level model (Model 3) for classifying subtypes of disease), the method comprising:
receiving one or more images of a body part of a patient (pg. 2 Methods: Dataset: images of 156 normal and 798 glaucomatous eyes were obtained);
inputting the one or more images into a diagnostic model (pg. 2-3 Proposed approach: In the first step, the model for classifying healthy and disease (disease detection) was built; build deep learning classification models with a single type of input image);
receiving, as output from the diagnostic model, a diagnosis for the patient (pg. 3 Experiments and training: We built a hierarchical classification model (Fig. 2b) applying transfer learning from the ImageNet-pretrained CNN model separately with a low-level model (Model 2 in Fig. 2b) for classifying normal versus glaucoma and a high-level model (Model 3 in Fig. 2b) for glaucoma classification (n = 4));
determining whether the diagnosis is positive for a given disease (pg. 3 Experiments and training: We built a hierarchical classification model (Fig. 2b) applying transfer learning from the ImageNet-pretrained CNN model separately with a low-level model (Model 2 in Fig. 2b) for classifying normal versus glaucoma and a high-level model (Model 3 in Fig. 2b) for glaucoma classification (n = 4); see further Fig. 2b Model 2 results in Normal or Disease);
responsive to determining that the diagnosis is positive, inputting a representation of the one or more images into a diagnosis subclassification model (Fig. 2b a high-level model (Model 3) for classifying subtypes of disease (e.g. when a disease is positive));
determining, based on output from the diagnosis subclassification model, a subclassification for the diagnosis (Fig. 2b Model 3 outputs a classification of the disease (e.g. Type-1, Type-2…Type-N); and
An fails to teach where Mavroeidis teaches outputting a control signal based on the subclassification (¶23 generating a control signal for modifying a graphical element based on the classification of the first sub-region of the image).
Therefore, it would have been obvious to one with ordinary skill in the art before the effective filing date of the invention to have implemented the teaching of outputting a control signal based on the subclassification from Mavroeidis into the method as disclosed by An. The motivation for doing this is to improve display information about a medical image of a subject.
Regarding claim 5, the combination of An and Mavroeidis disclose the method of claim 1, wherein the diagnosis subclassification model is trained using a training set comprising historical representations of medical images as labeled with one or more subclassifications (An pg. 3 Experiments and training: The first experiment was performed as described below by using the entire set of training data to compare classification performances among models trained using three proposed approaches (Fig. 2). We separately trained deep learning models for each type of extracted image from the 3D OCT data. A convolutional neural network (CNN), which can automatically create efficient image features for the classification, was used as the classifier27. Flat classification models were used to classify eyes as normal, FI, GE, MY, or SS directly with transfer learning from a deep learning model (a CNN) pretrained on the ImageNet dataset (ImageNet-pretrained CNN model) to create Model 1 (Fig. 2a, n = 4). We built a hierarchical classification model (Fig. 2b) applying transfer learning from the ImageNet-pretrained CNN model separately with a low-level model (Model 2 in Fig. 2b) for classifying normal versus glaucoma and a high-level model (Model 3 in Fig. 2b) for glaucoma classification (n = 4). Furthermore, the normal confidence from Model 2 and the confidence of FI, GE, MY, and SS from Model 3 were concatenated into a confidence vector length of 5. Then, a metamodel of the linear support vector machine (SVM) was trained using the confidence vector data with the supervised labels to combine the models in a cascaded manner).
Regarding claim 7, the combination of An and Mavroeidis disclose the method of claim 1, wherein the control signal is determined from a plurality of candidate control signals based on the subclassification (Mavroeidis ¶23 generating a control signal for modifying a graphical element based on the classification of the first sub-region of the image. The graphical element may then be displayed in accordance with the control signal; ¶112 to generate a display control signal for modifying at least one of the size, shape, position, orientation, pulsation or colour of the graphical element based on the image sub-region classifications).The motivation to combine the references is discussed above in the rejection for claim 1.
Regarding claim(s) 8, 12 and 14 (drawn to a CRM):
The rejection/proposed combination of An and Mavroeidis, explained in the rejection of method claim(s) 1, 5 and 7, anticipates/renders obvious the steps of the computer readable medium of claim(s) 8, 12 and 14 because these steps occur in the operation of the proposed combination as discussed above. Thus, the arguments similar to that presented above for claim(s) 1, 5 and 7 is/are equally applicable to claim(s) 8, 12 and 14. See further Mavroeidis ¶126-131.
Regarding claim(s) 15 and 19 (drawn to a system):
The rejection/proposed combination of An and Mavroeidis, explained in the rejection of method claim(s) 1 and 5, anticipates/renders obvious the steps of the system of claim(s) 15 and 19 because these steps occur in the operation of the proposed combination as discussed above. Thus, the arguments similar to that presented above for claim(s) 1 and 5 is/are equally applicable to claim(s) 15 and 19. See further Mavroeidis ¶126-131.
Claim(s) 2, 4, 9, 11, 16, and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over the combination of An and Mavroeidis as applied to claims 1, 8, and 15 above, and further in view of Abramoff et al (US 20200037930).
Regarding claim 2, the combination of An and Mavroeidis disclose the method of claim 1, but fail to teach where Abramoff teaches wherein the diagnostic model comprises an extraction model and a diagnosis model (¶16 Where the images are sufficient for identifying image biomarkers, image analysis service 130 extracts the image biomarkers and transmits the image biomarkers to classifier service 160) and produces the diagnosis by: inputting the one or more images into the extraction model (¶32 Image biomarker extraction module 233 extracts image biomarkers from the images received from image capture apparatus 120); receiving, as output from the extraction model, biomarkers indicative of disease (¶32 Image biomarker extraction module 233 extracts image biomarkers from the images received from image capture apparatus 120. Image biomarkers include the detection of relevant anatomy and disease/pathological features associated with the diagnosis. Image biomarker extraction module 233 may determine what attributes of an image form a biomarker based on entries of image biomarker attributes database 235); inputting the biomarkers into the diagnosis model (¶21 Classifier service 160 also receives image biomarkers from image analysis service 130 and/or acoustic biomarkers from acoustic response analysis service 150); and receiving indicia of the diagnosis from the diagnosis model (¶22 Classifier service 160 synthesizes the received data and feeds the synthesized data as input into a machine learning model. The machine learning model outputs a probability-based diagnosis, which classifier service 160 transmits to therapy determination service 170.).
Therefore, it would have been obvious to one with ordinary skill in the art before the effective filing date of the invention to have implemented the teaching of wherein the diagnostic model comprises an extraction model and a diagnosis model and produces the diagnosis by: inputting the one or more images into the extraction model; receiving, as output from the extraction model, biomarkers indicative of disease; inputting the biomarkers into the diagnosis model; and receiving indicia of the diagnosis from the diagnosis model from Abramoff into the method as disclosed by the combination of An and Mavroeidis. The motivation for doing this is to improve diagnosing diseases in a patient.
Regarding claim 4, the combination of An and Mavroeidis disclose the method of claim 1, but fails to teach where Abramoff teaches wherein the representation of the one or more images that is input into the diagnosis subclassification model comprises biomarkers extracted from the one or more images by an extraction model (¶21 Classifier service 160 also receives image biomarkers from image analysis service 130 and/or acoustic biomarkers from acoustic response analysis service 150).
Therefore, it would have been obvious to one with ordinary skill in the art before the effective filing date of the invention to have implemented the teaching of wherein the representation of the one or more images that is input into the diagnosis subclassification model comprises biomarkers extracted from the one or more images by an extraction model from Abramoff into the method as disclosed by the combination of An and Mavroeidis. The motivation for doing this is to improve diagnosing diseases in a patient.
Regarding claim(s) 9 and 11 (drawn to a CRM):
The rejection/proposed combination of An, Mavroeidis, and Abramoff, explained in the rejection of method claim(s) 2 and 4, anticipates/renders obvious the steps of the computer readable medium of claim(s) 8 and 11 because these steps occur in the operation of the proposed combination as discussed above. Thus, the arguments similar to that presented above for claim(s) 2 and 4 is/are equally applicable to claim(s) 9 and 11. See further Mavroeidis ¶126-131.
Regarding claim(s) 16 and 18 (drawn to a system):
The rejection/proposed combination of An, Mavroeidis, and Abramoff, explained in the rejection of method claim(s) 2 and 4, anticipates/renders obvious the steps of the system of claim(s) 16 and 18 because these steps occur in the operation of the proposed combination as discussed above. Thus, the arguments similar to that presented above for claim(s) 2 and 4 is/are equally applicable to claim(s) 16 and 18. See further Mavroeidis ¶126-131.
Claim(s) 3, 10, and 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over the combination of An, Mavroeidis and Abramoff as applied to claim 2, 9, and 16 above, and further in view of Patil et al (US 20240194342).
Regarding claim 3, the combination of An, Mavroeidis, and Abramoff disclose the method of claim 2, but fail to teach where Patil teaches wherein determining whether the diagnosis is positive for a given disease comprises determining whether the diagnosis matches one of a plurality of predefined diseases for which subclassification is eligible (¶99 Referring to FIGS. 5A and 5C, in some embodiments, the plurality of predefined medical conditions 306 include one or more of: asthma, heart disease, stroke, diabetes, arthritis, cancer, obesity, Alzheimer's disease, substance abuse, influenza, HIV, a Zoonotic disease, tuberculosis, a chronic kidney disease, and mental illness; each of the first subset of the sub-classifier models 302-1 to 302-N includes a single-class classifier 400, and a value of a single output PB of the classifier 400 (e.g., condition data 502 in FIG. 5A) is binary (e.g., equal to “0” or “1”) indicating whether a predefined medical condition 306 is identified.).
Therefore, it would have been obvious to one with ordinary skill in the art before the effective filing date of the invention to have implemented the teaching of wherein determining whether the diagnosis is positive for a given disease comprises determining whether the diagnosis matches one of a plurality of predefined diseases for which subclassification is eligible from Patil into the method as disclosed by the combination of An, Mavroeidis, and Abramoff. The motivation for doing this is to improve integrating medical data from a plurality of distinct modalities to provide supplement medical information and determine actionable insights.
Regarding claim(s) 10 (drawn to a CRM):
The rejection/proposed combination of An, Mavroeidis, Abramoff, and Patil explained in the rejection of method claim(s) 3, anticipates/renders obvious the steps of the computer readable medium of claim(s) 10 because these steps occur in the operation of the proposed combination as discussed above. Thus, the arguments similar to that presented above for claim(s) 3 is/are equally applicable to claim(s) 10. See further Mavroeidis ¶126-131.
Regarding claim(s) 17 drawn to a system):
The rejection/proposed combination of An, Mavroeidis, Abramoff, and Patil, explained in the rejection of method claim(s) 3, anticipates/renders obvious the steps of the system of claim(s) 17 because these steps occur in the operation of the proposed combination as discussed above. Thus, the arguments similar to that presented above for claim(s) 3 is/are equally applicable to claim(s) 17. See further Mavroeidis ¶126-131.
Claim(s) 6, 13, and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over the combination of An and Mavroeidis as applied to claim 1, 8, and 15 above, and further in view of Hagendorn et al (US 9754383).
Regarding claim 6, the combination of An and Mavroeidis disclose the method of claim 1, but fail to teach where Hagendorn teaches wherein determining the subclassification comprises inputting the output from the diagnosis subclassification model into a rules engine, the rules engine determining the subclassification (col 6 lines 27-35 The stratification scheme entails applying selection criteria (i.e. thresholds, logical operators, etc.) to the disease severity score based on the SA:V value. The stratification scheme bins individuals submitted for evaluation into two diagnostic bins for celiac disease (i.e. positive diagnosis of celiac disease and negative diagnosis of celiac disease bins) and further sub-classifies the positive diagnosis bin into two or more sub-bins corresponding to disease severity (i.e. Marsh scores of 1, 3A, 3B, 3C).).
Therefore, it would have been obvious to one with ordinary skill in the art before the effective filing date of the invention to have implemented the teaching of wherein determining the subclassification comprises inputting the output from the diagnosis subclassification model into a rules engine, the rules engine determining the subclassification from Hagendorn into the method as disclosed by the combination of An and Mavroeidis. The motivation for doing this is to improve automated methods for assessing tissue morphometry in digital images.
Regarding claim(s) 13 (drawn to a CRM):
The rejection/proposed combination of An, Mavroeidis, and Hagendorn explained in the rejection of method claim(s) 6, anticipates/renders obvious the steps of the computer readable medium of claim(s) 13 because these steps occur in the operation of the proposed combination as discussed above. Thus, the arguments similar to that presented above for claim(s) 6 is/are equally applicable to claim(s) 13. See further Mavroeidis ¶126-131.
Regarding claim(s) 20 drawn to a system):
The rejection/proposed combination of An, Mavroeidis, and Hagendorn, explained in the rejection of method claim(s) 6, anticipates/renders obvious the steps of the system of claim(s) 20 because these steps occur in the operation of the proposed combination as discussed above. Thus, the arguments similar to that presented above for claim(s) 6 is/are equally applicable to claim(s) 20. See further Mavroeidis ¶126-131.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to KEVIN KY whose telephone number is (571)272-7648. The examiner can normally be reached Monday-Friday 9-5PM.
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, Vincent Rudolph can be reached at 571-272-8243. 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.
/KEVIN KY/ Primary Examiner, Art Unit 2671
Prosecution Insights