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 filed 02/12/2026 fails to comply with 37 CFR 1.98(a)(2), which requires a legible copy of each non-patent literature publication or that portion which caused it to be listed. Specifically, Rasheed et al A comprehensive experiment-based review of low-light image enhancement methods and benchmarking low-light image quality assessment (Hereinafter, “Rasheed”) has not been made fully available to the examiner (the examiner assumes an extraction error resulted in only a small portion of the document (a part of the title) being shown). The information referred to within Rasheed has therefore not been considered.
Claim Interpretation
The claims will be read under the broadest reasonable interpretation standard outlined in
MPEP § 2111.01.
As a note, claims 1 and 11 recite a “large-scale hierarchical image database”. So as to avoid issuing an objection or rejection for relative terminology, this phrase must be interpreted within its understood technical meaning. Accordingly, a “large-scale hierarchical image database” is narrowly understood to refer to ImageNet and any other direct equivalents.
Specification
The abstract of the claimed invention is objected to for the following informality:
“early recognition of squamous cell carcinoma in epidermolysis bullosa” reads as a typographical error for “early recognition of squamous cell carcinoma in patients suffering from epidermolysis bullosa”.
The same error is noted on page 1 of the claimed invention’s specification.
Appropriate corrected is required.
Claim Objections
Claim 1 is objected to for the following informality:
“combining output of the del trained” reads as a typographical error for “combining output of the model trained”.
Claim 3 is objected to for the following informality:
“trail the AI tool” reads as a typographical error for “train the AI tool”.
Claim 11 is objected to for the following informality:
The claim lacks a transitional preamble, such as “the method comprising:”
Lines 2-3, “epidermolysis bullosa” should be changed to “epidermolysis bullosa (EB)”
Claim 15 is objected to for the following informality:
“determining whether the lesion is an SCC lesion of a non-SCC lesion” reads as a typographical error for “determining whether the lesion is an SCC lesion or a non-SCC lesion”.
Claim 16 is objected to for the following informality:
The claim recites the administration of treatment for a skin condition. However, it depends on claim 11, which is “a method of developing an artificial intelligence (AI) tool…”. The examiner assumes that this claim was meant to depend on claim 15.
Appropriate correction is required.
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.
Claims 1-16 are rejected under 35 U.S.C. 112(b) as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention.
With respect to claims 1 and 11, the term “robust” in claim 1 is a relative term which renders the claim indefinite. The term “robust” is not defined by the claim. The non-limiting examples of the specification do not provide a sufficient standard for ascertaining the requisite degree, and accordingly one of ordinary skill in the art would not be reasonably apprised of the scope of the invention. While it is established that the “initially trained model” will possess “non-specific feature-detection capabilities”, such models can naturally vary significantly in overall efficiency. ResNet50 and ResNet152 are described as “robust” in page 17 of the claimed invention’s specification, but as the claim implicates a wider variety of “initially trained model[s]”, further clarification is necessary. The examiner otherwise suggests simply removing the term.
Claims 2-10, 12-14, and 16 are rejected by virtue of their dependency on claims 1 and 11.
With respect to claim 10, an additional issue of antecedent basis is noted. The claim introduces a “first” or “second” medical condition, but step (a) refers back to “the medical condition”. As such, it is unclear which medical condition is displayed. The examiner suggests changing the language to “displaying a symptom of the first or second medical conditions”, or other equivalents.
With respect to claims 13 and 15, the same antecedent basis issue is noted as with claim 10, instead with respect to “an SCC lesion or a non-SCC lesion” and “the lesion”.
Claim 14 is additionally rejected by virtue of its dependency on claim 13.
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 1-16 are rejected under 35 U.S.C. § 101 because they are directed to ineligible patent subject matter. The claims are directed to the Abstract Idea groupings of mental processes under MPEP § 2106.04(a)(2)(III) and mathematical calculations under MPEP § 2106.04(a)(2)(I). This is a judicial exception under Step 2A, Prong One of the framework established by the cases of Alice Corp. Pty. Ltd. v. CLS Bank Int'l, 573 U.S. 208, 216, 110 USPQ2d 1976, 1980 (2014) and Mayo Collaborative Servs. v. Prometheus Labs., Inc., 566 U.S. 66, 71, 101 USPQ2d 1961, 1965 (2012). See MPEP § 2106.04(II).
PNG
media_image1.png
200
400
media_image1.png
Greyscale
Step 1: The claims in question are directed to methods for “early recognition of rare skin diseases”. Methods (processes) are a statutory category. See MPEP 2106.03(I), “NTP, Inc. v. Research in Motion, Ltd., 418 F.3d 1282, 1316, 75 USPQ2d 1763, 1791 (Fed. Cir. 2005) ("[A] process is a series of acts.") (quoting Minton v. Natl. Ass’n. of Securities Dealers, 336 F.3d 1373, 1378, 67 USPQ2d 1614, 1681 (Fed. Cir. 2003)). As defined in 35 U.S.C. 100(b), the term "process" is synonymous with "method."” (Step 1: Yes).
Step 2A, Prong One: As explained in MPEP 2106.04(II), a claim “recites” a judicial
exception when the judicial exception is “set forth” or “described” in the claim. Here, each
claim recites or depends upon the mental processes of image augmentation, training, adapting, combining output, and training a classifier (Claim 1, “training the AI tool…adapting the initially trained model to the medical conditions by transfer learning…combining output…training a Random Forest-based meta-learner classifier”; Claim 3, “further comprising applying augmentation techniques…”; Claim 11), using the AI tool trained (Claim 10, “obtaining an image…providing the image…determining whether the subject suffers from the first or second condition”; Claim 13), and administering treatment (Claim 14; Claim 16). The claim recites processes which implicate mathematical calculations as well (Claim 1, “adapting the initially trained model to the medical conditions by transfer learning…combining output…a random forest-based meta-learner…”; Claim 11)
The claims are recited at a high level of generality and lack any specifics precluding such
an analysis from being interpreted under the mental processes grouping of “practically performed
in the mind” (see also MPEP § 2106.04(a)(2) identifying how e.g. a use of pen and paper, a ruler,
or a computer as a tool (to assist in visually/mentally analyzing/observing acquired
images/video) fails to preclude such an interpretation under the mental processes judicial
exception). Activities which generally take place in digital space may therefore be performed mentally, even if they may require the additional computer tool. Similarly, basic weight adjustment, image augmentation, machine-learning calculation components, and treatment administration do not elevate these claims past a mental process/mathematical calculation.
Regarding artificial intelligence, to the extent it is implicated, the claims are comparable to Claim 2 of Example 47 of the July 2024 PEG regarding subject matter eligibility (https://www.uspto.gov/sites/default/files/documents/2024-AISMEUpdateExamples47-49.pdf). As stated therein, an artificial intelligence’s analyses, detections, and reinforcement learnings may be practically performed in the human mind. To the extent mathematical calculations are required to operate and train the artificial intelligence in image analysis, the separate judicial exception is also implicated.
As such, the usage of a computer to determine whether an image shows a medical condition, does not elevate these claims beyond a mental process/mathematical calculation. (Step 2A, Prong One: Yes).
Step 2A, Prong Two: If Prong One of Step 2A is met, the examiner must consider (1)
whether there are any ‘additional elements’ recited in the claim beyond the judicial exception,
and (2) evaluate those additional elements individually and in combination to determine whether
the claim as a whole integrates the exception into a practical application. See MPEP §
2106.04(d).
Limitations the courts have found indicative of integration include: an improvement in
the functioning of a computer, or an improvement to other technology or technical field, as
discussed in MPEP §§ 2106.04(d)(1) and 2106.05(a); applying or using a judicial exception to
effect a particular treatment or prophylaxis for a disease or medical condition, as discussed in
MPEP § 2106.04(d)(2) (the examiner notes that the factors of MPEP § 2106.04(d)(2) are not met. Claims 14 and 16 are the closest to satisfying the factors, but still fail for generality. MPEP § 2106.04(d)(2)(a), “Conversely, consider a claim that recites the same abstract idea and "administering a suitable medication to a patient." This administration step is not particular, and is instead merely instructions to "apply" the exception in a generic way. Thus, the administration step does not integrate the mental analysis step into a practical application”); implementing a judicial exception with, or using a judicial exception in
conjunction with, a particular machine or manufacture that is integral to the claim, as discussed
in MPEP § 2106.05(b); effecting a transformation or reduction of a particular article to a
different state or thing, as discussed in MPEP § 2106.05(c); and applying or using 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, as discussed in MPEP § 2106.05(e).
Limitations that the courts have found non-indicative of integration include: merely
reciting the words "apply it" (or an equivalent) with the judicial exception, or merely including
instructions to implement an abstract idea on a computer, or merely using a computer as a tool to
perform an abstract idea, as discussed in MPEP § 2106.05(f); adding insignificant extra-solution
activity to the judicial exception, as discussed in MPEP § 2106.05(g); and generally linking the
use of a judicial exception to a particular technological environment or field of use, as discussed
in MPEP § 2106.05(h).
As an additional note, ‘additional elements’ are generally limitations excluded from
interpretation under the Abstract Idea groupings, and may comprise portions of limitations
otherwise identified as falling under those Abstract Idea groupings of the 2019 PEG (e.g. any
‘determination’ that may be made mentally by a user, neural network and/or generic computer hardware is considered under the ‘apply it’ considerations of 2106.05(f)). Any ‘providing’/outputting broadly, and ‘collection/input’ of data (i.e. input of training/evaluation images, output of determination result), also fail(s) to integrate at least in view of MPEP 2106.05(g) (extra-solution data gathering/output) and/or 2106.05(h) as ‘generally linking’ the exception to a field of use involving machine learning and/or imagery so acquired (e.g. the use of a camera/computer to acquire said image broadly). The same determination holds for dependent claims that serve to limit the collection/output of data/images (by means of what is collected based on recited conditions) and/or introduce limitations generally linking to a field of use.
None of the instant claims appear to explicitly/clearly capture/recite any disclosed
improvement in technology (see MPEP 2106.05(a), with note that ‘functioning of a computer’
concerns functions integral to the way a computer operates and not ‘functions’ that a generic
computer can be programmed/adapted to perform (see also 2106.05(f))) and any ‘additional
elements’, even when considered in combination, fail to integrate at Prong Two of Step 2A
accordingly. Integration in view of subsection (a) requires an identification of the manner in
which the improvement is achieved, to be explicitly and specifically recited in the claims, as
‘additional elements’ precluded from interpretation under any of the Abstract Idea groupings
(since the improvement cannot be to the exception itself). With reference to MPEP 2106.05(a):
It is important to note, the judicial exception alone cannot provide the improvement. The improvement can be provided by one or more additional elements. See the discussion of Diamond v. Diehr, 450 U.S. 175, 187 and 191-92, 209 USPQ 1, 10 (1981))
Claim 1 recites the “AI tool” as a subject to be trained; Claims 10, 13, and 15 recite the “AI tool” as a component of the broader method. The language as written implicates MPEP § 2106.05(f)(1), “Mere Instructions to Apply an Exception” – “Whether the claim recites only the idea of a solution or outcome i.e., the claim fails to recite details of how a solution to a problem is accomplished”. A general outline is provided for training (and using) an AI tool, but no significant details of underlying analysis are given beyond the conclusory level. For example, the claim recites “combining output of the model trained on the two medical conditions with patient clinical data to generate a combined model”, but no detail is provided as to how the (presumably non-image) data is combined. Similarly, the tool is described as being largely composed of “models”, which implicates no specific architecture beyond how the models are broadly combined. These concerns mirror claim 1 of Example 48 of the July 2024 PEG provided above, specifically the analysis at Step 2A, Prong Two on page 20. Eligibility would closer resemble claim 3 of Example 48 (page 25), which instead clarifies the precise means by which the DNN (or in this case, the “AI tool”) operates.
Even when viewed in combination, any additional elements present do not integrate the
recited judicial exceptions into a practical application (Step 2A, Prong Two: No), and the claims are directed to the judicial exceptions (Revised Step 2A: Yes → Step 2B).
Step 2B: If Prong Two of Step 2A is not met, the examiner must consider whether the
claim as a whole amounts to ‘significantly more’ than the recited exception, i.e., whether any
‘additional element’, or combination of additional elements, adds an inventive concept to the
claim. The considerations of Step 2A Prong 2 and Step 2B overlap, but differ in that 2B also
requires considering whether the claims feature any “specific limitation(s) other than what is
well-understood, routine, conventional activity in the field” (WURC) (MPEP § 2106.05(d)).
Such a limitation if specifically recited however, must still be excluded from interpretation under
any of the Abstract Idea groupings. Step 2B further requires a re-evaluation of any additional
elements drawn to extra-solution activity in Step 2A (e.g. obtaining images) – however no limitations appear directed to any novel collection per se. Limitations not indicative of an inventive concept/‘significantly more’ include those that are not specifically recited (instead recited at a high level of generality), those that are established as WURC (a plurality of cited references serve to evidence the WURC nature of ‘analysis’ based at least in part on corroborating/additional ground data), and/or those that are not ‘additional elements’ by nature of their analysis at Prong One of Step 2A (i.e. directed to the exception – see above re. deciding that a second acquisition may be advantageous/desired). The July 2024 PEG describes that an improvement/ inventive concept (for ‘significantly more’ determination(s)) cannot be to the judicial exception itself. As additionally recited by page 26 of the specification of the claimed invention, non-limiting examples of the “models” are understood to encompass a plurality of WURC machine-learning models, including “ResNet50, ResNet101, ResNet 152…Xception…VGG16…DenseNet121…Vision Transformer…”.
The claims in question recite little beyond those limitations recited at a high level
of generality and falling under e.g. the mental processes/mathematical calculation Abstract Idea groupings, and would monopolize the exceptions accordingly. The additional limitations of computer processing, machine-learning, and user interface as recited are WURC, as further evidenced by the body of prior art cited by the examiner in this office action. (Step 2B: No).
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.
Claims 1-7 and 10 are rejected under 35 U.S.C. 103 as being unpatentable over Nunnari et. al A Study on the Fusion of Pixels and Patient Metadata in CNN-Based Classification of Skin Lesion Images (Hereinafter, “Nunnari”) in view of Qureshi et. al Transfer Learning with Ensembles of Deep Neural Networks for Skin Cancer Detection in Imbalanced Data Sets (Hereinafter, “Qureshi”)
With respect to claim 1, Nunnari teaches:
A method of developing an artificial intelligence (AI) tool for distinguishing between two medical conditions (Fig. 4), the method comprising:
training the AI took on a large-scale hierarchical image database to generate an initially trained model with robust non-specific feature-detection capabilities ([4.1] “We follow a transfer learning approach by first initializing the weights of the original VGG16/RESNET50 of a pre-trained model from ImageNet [6] and then substituting the final part of the architecture”)
adapting the initially trained model to the medical conditions by transfer learning to [4.1] “We follow a transfer learning approach by first initializing the weights of the original VGG16/RESNET50 of a pre-trained model from ImageNet and then substituting the final part of the architecture”; [4] “Figure 4 depicts the methods used for integrating metadata together with pixel-based image classification. The pipeline starts with an initial training of a deep CNN for the classification of an image across the eight ISIC 2019 categories”; examiner notes that “comprising” only requires that “the two medical conditions” be included amongst the plurality of medical conditions)
combining the output of the model trained on the two medical conditions with patient clinical data to generate a combined model (Fig. 4)
training a Random Forest-based meta-learner classifier with the combined model to generate the AI tool for distinguishing between two medical conditions ([4.3] “In this approach, we concatenate the final probability density from our baseline networks with the metadata information… We experimented with several well-established Machine Learning algorithms like Support Vector Machines (SVM), Gradient Boosting and Random Forest…Finally, we use the ensemble learning method of Random Forests which is a decision tree algorithm that trains multiple trees and, for classification problems, picks the class with the highest mode during testing”)
Nunnari does not explicitly teach:
binary classification
However, Qureshi, in the same field of endeavor of skin cancer detection, teaches:
A method of developing an artificial intelligence (AI) tool for distinguishing between two medical conditions (Fig. 1), the method comprising:
training the AI took on a large-scale hierarchical image database to generate an initially trained model with robust non-specific feature-detection capabilities ([2] “The base learners are either pre-trained on balanced dataset collected from ISIC archive or on the the ISIC 2020 dataset.”; however with understanding that ImageNet pre-training is acknowledged in other methods, [1] “Hosny’s technique is based on a pre-trained AlexNet network (a specific deep learning architecture proposed by Krizhevsky et al. (2012)) that was originally trained to classify images on a commonly used ImageNet dataset, and then adapted to perform skin cancer classification by transfer learning”)
adapting the initially trained model to the medical conditions by transfer learning ([2.2] “Transfer learning during training of the base-learners”) to binary classification of images of the two medical conditions to generate model trained on the two medical conditions (Fig. 1; [2]; [3.1] “All the images are labelled using histopathology and expert opinion either as benign or malignant skin lesions”; Appendix A, showing single neuron sigmoid activation layer of the base learners outputting a percentage along the binary scale of malignant/benign; Table 3, directly comparing classification result of base learners to trained SVM meta-classifier, further evidencing that base learners are also configured to render the same binary classification output)
combining the output of the model trained on the two medical conditions with patient clinical data to generate a combined model (Fig. 1, “Auxiliary data”; [2] “Predictions from all the base learners along with the auxiliary data contained in the metadata associated to the images are used as input to an SVM classifier, which finally classifies each image in dataset as positive (malignant) or negative (benign)”; Table 1: Metadata)
training a [1] “In the second step, all the predictions from base learners along with the meta-data, including, e.g., the age and gender of the subject, provided with the input images is provided to an SVM meta-learner to obtain the final classification”; Table 3, “SVM meta-classifier”) with the combined model to generate the AI tool for distinguishing between two medical conditions (Fig. 1; though acknowledging random forest classifiers may be used in place of SVM classifiers in similar tasks; Table 2; [1] and Fig. 2, showing generally that “benign” images may also reasonably include other medical conditions)
PNG
media_image2.png
685
630
media_image2.png
Greyscale
It would have been obvious to one of ordinary skill in the art as of the effective filing date of the claimed invention to modify Nunnari to include the limitations of binary classification, as taught by Qureshi. Doing so would provide a simpler output, and further reduce the computational complexity of the system. The systems readily integrate. A system which is configured to distinguish between a variety of benign and malignant skin conditions could predictably also output a simpler determination of “benign” vs “malignant”.
With respect to claim 2, Nunnari/Qureshi teach:
The method of claim 1, wherein the large-scale hierarchical image database is not specific to images of the two medical conditions (Nunnari, [4.1] “ImageNet”, directly citing to Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., Fei-Fei, L.: ImageNet: large-scale hierarchical image database. In: 2009 IEEE Conference on Computer Vision and Pattern Recognition, pp. 248–255. IEEE, Miami, June 2009)
PNG
media_image3.png
477
1432
media_image3.png
Greyscale
With respect to claim 3, Nunnari/Qureshi teach:
The method of claim 1, further comprising applying augmentation techniques to expand diversity of the images of the two medical conditions used to train the AI tool (Nunnari, [4.1] “We trained our VGG16/RESNET50 baseline models for 10/30 epochs, batch size 8/32, SGD optimizer, lr = 1E-5, with a 48 augmentation factor (each image is flipped and rotated 24 times at 15 steps, as in Fujisawa et al. [10]). Training takes about 3/5 days on an NVIDIA RTX 2080Ti GPU”)
With respect to claim 4, Nunnari/Qureshi teach:
The method of claim 3, where augmentation techniques comprise transformations of the images of the two medical condition used to train the AI tool (Nunnari, [4.1])
With respect to claim 5, Nunnari/Qureshi teach:
The method of claim 4, wherein transformations of the images comprise one or more of rotations, flips, scaling, and color adjustments (Nunnari, [4.1])
With respect to claim 6, Nunnari/Qureshi teach:
The method of claim 1, wherein the two medical conditions are skin conditions (Nunnari, Fig. 4, “Images (ISIC2019)”; Qureshi, [3.1])
With respect to claim 7, Nunnari/Qureshi teach:
The method of claim 1, wherein the two medical conditions can be differentiated visually (Nunnari, Fig. 2; Qureshi, Fig. 2)
With respect to claim 10, Nunnari/Qureshi teach:
A method of determining whether a subject suffers from a first or second medical condition, the method comprising:
obtaining an image of the subject displaying a symptom of the medical condition (Nunnari, Fig. 4)
providing the image to an AI tool developed by the method of claim 1 (Nunnari, Fig. 4; Nunnari, [5, Results]; Qureshi, [4, Experimental Results])
determining whether the subject suffers from the first or second condition (Nunnari, [4] “The fusion with metadata is then performed in two ways. In the first fusion method (non-NN, Sect. 4.3), the output of the CNN classification (softmax, size 8) is concatenated with the metadata. The resulting feature vector (size 19) is passed to several well-known learning methods (either tree-based or SVM). In the second fusion method, (sNN, Sect. 4.4), we take, from the CNN, either the softmax output (size 8), or the activation values of the two fully connected layers (size 2048, each) located between the last convolution stage and the final softmax. Each fusion vector is then the concatenation of the softmax output (or the activation values for that matter) and the metadata vector. The concatenated vector is passed through a shallow neural network of two or three layers.”; one of the 8 categories, necessarily including the “first” and “second” condition; Nunnari, [4.3] “Finally, we use the ensemble learning method of Random Forests which is a decision tree algorithm that trains multiple trees and, for classification problems, picks the class with the highest mode during testing”)
Claim 15 is rejected under 35 U.S.C. 103 as being unpatentable over Krupiczojc et. al Management of cutaneous squamous cell carcinoma in patients with epidermolysis bullosa (Hereinafter, “Krupiczojc”) in view of Gabashvili ChatGPT in Dermatology: A comprehensive Systematic Review (Hereinafter, “Gabashvili”) and Yang et. al MM-REACT: Prompting ChatGPT for MultiModal Reasoning and Action (Hereinafter, “Yang”)
With respect to claim 15, Krupizocjc teaches:
A method of determining whether a lesion on a subject suffering from epidermolysis bullosa (EB) is a squamous cell carcinoma (SCC) lesion or a non-SCC lesion ([Abstract]), the method comprising
obtaining [Abtract])
[Abstract] “In this issue of the BJD, an international multidisciplinary group of EB experts provide for the first time comprehensive guidelines on the diagnosis and treatment of SCC in individuals with EB…Firstly, the authors recommend full-skin examinations every 3–6 months from as early as 10 years of age in the RDEB patient group and from 20 years of age in the lower-risk groups. A detailed list of clinical indicators of SCC is provided, which is of value to the clinician as identification of SCC in chronic wounds is often challenging”)
determining whether the lesion is an SCC lesion or a non-SCC lesion ([Abstract])
Krupizocjc does not explicitly teach:
obtaining an image of the lesion
providing the image to an AI tool trained to distinguish between SCC lesions and non-SCC lesions on a subject suffering from EB
However, Gabashvili teaches:
obtaining an image of Page 10-11)
providing the image to an AI tool Page 10-11, “For dermatology diagnostics, SkinGPT [25] presents an innovative system that combines MiniGPT-4, an advanced vision-based large language model, with an extensive collection of in-house skin disease images and doctor’s notes. Unlike generic ChatGPT, SkinGPT can be locally deployed and effectively handle sensitive images”)
It would have been obvious to one of ordinary skill in the art as of the effective filing date of the claimed invention to modify Krupizocjc to include the limitations of AI identification, as taught by Gabashvili. Doing so would allow a dermatologist (in the process of examining a given lesion for an EB patient) to capture an image, and then leverage AI image analysis and efficiency to further determine if a skin condition is present. The systems readily integrate, as visual examination is already a known feature of skin diagnostics generally.
Krupizocjc/Gabashvili do not explicitly teach:
trained to distinguish between SCC lesions and non-SCC lesions on a subject suffering from EB
However, Yang, in the same field of endeavor of AI informatics, teaches:
trained to distinguish between SCC lesions and non-SCC lesions on a subject suffering from EB (Fig. 1; Fig. 3)
The examiner notes that the claim requires no specific architecture or standard of efficiency. The claim merely requires that the AI tool be trained to distinguish between two conditions, even if it does so poorly. As of the effective filing date of the claimed invention, it was possible for one of ordinary skill in the art to pass images of the claimed skin conditions to a general LLM. The LLM, in its vision-language capacity, would then attempt to answer the question using its general protocols (such as by searching the internet), which would demand it to find or at least predict an answer. Therefore, at this high-level, one could say a general LLM (which is trained to predict the next token and consult the internet when there are gaps in its knowledge) is an AI tool “trained to distinguish between SCC lesions and non-SCC lesions on a subject suffering from EB”, even if it does not do so well.
It would have been obvious to one of ordinary skill in the art as of the effective filing date of the claimed invention to modify Krupizocjc/Gabashvili to include the limitations of AI training, as taught by Yang. Doing so would provide a broader corpus of information to draw from and predictions to ultimately make regarding skin conditions. The systems readily integrate, as Krupizocjc/Gabashvili already draws from the advantages of large language models.
Allowable Subject Matter
Claims 8, 9, 11-14, and 16 overcome the prior art as written, though otherwise remain rejected under 35 U.S.C. § 112(b) and 35 U.S.C. § 101.
Claims 8, 9, 11-14, and 16 explicitly distinguish between an SCC lesion and a non-SCC lesion in subjects suffering from epidermolysis bullosa, within their overarching structure. Even if one takes a broad interpretation of the claims, to say that the methods need not pay attention to underlying EB conditions (and could simply distinguish between the lesions blindly), the cited references and their respective databases do not explicitly include patients with the required combination of EB and SCC/non-SCC lesions. As of the effective filing date of the claimed invention, such an image database was not readily accessible (further reflected in pages 12-13 of the claimed invention’s specification). To the extent an additional database could be found and/or combined to arrive at the language of the claims, there would still lack sufficient motivation to one of ordinary skill in the art to modify the references of record. The methods of the cited art are indifferent to epidermolysis bullosa.
Similarly, the prior art does not teach an artificial intelligence tool which explicitly distinguishes between SCC and non-SCC lesions in EB patients in the required specificity. Given the unique challenges associated with the claimed invention’s EB/SCC-specific solution, the obviousness of combining any existing AI architectures would face difficulties. Accordingly, the claims overcome the prior art as written.
Additional References
Additionally cited references (see attached PTO-892) otherwise not relied upon above have been made of record in view of the manner in which they evidence the general state of the art. In particular, the examiner notes Tevaramani et. al Skin Cancer Classification Using Convolution Neural Network and Meta Learning, which teaches a combination model followed by a meta-learner classifier (Fig. 1; [5.2]); and Ma et. al EFFNet: A skin cancer classification model based on feature fusion and random forests, which teaches transfer-learning using a model pre-trained on the ImageNet dataset, as well as image augmentation (Page 2; Fig. 1).
Inquiry
Any inquiry concerning this communication or earlier communications from the examiner should be directed to NOAH WILLIAM BOYAR whose telephone number is (571)272-8392. The examiner can normally be reached 8:30 – 5:00 EST, Monday – Friday.
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, Chan Park can be reached at 571-272-7409. 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.
/NOAH W BOYAR/Examiner, Art Unit 2669
/CHAN S PARK/Supervisory Patent Examiner, Art Unit 2669
Prosecution Insights