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 .
Preliminary Amendment
In response to applicant’s preliminary amendment received on 12/19/2024, all requested changes to the claims and specification have been entered.
Claim(s) 1-48 were previously pending.
No Claim(s) have been added.
Claim(s) 4, 8, 13, 17, 22, 25-26 & 28-48 have been cancelled.
Claim(s) 1-3, 5-7, 9-12, 14-16, 18-21, 23-24 & 27 are currently pending.
Priority
Acknowledgement is made of applicant’s claim for priority under 35 U.S.C. 119(e) to US provisional application, 63/353,714, filed 06/02/2022.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 12/19/2024 is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 1-3, 5-7, 10-12, 14-16, 18-21, 23, 24 & 27 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claims 1, 10 & 19 introduce the limitation “extract, with respect to each of said subjects, a set of features representing said images and said datapoints associated with said subject”. It is unclear what exactly constitutes a feature and what the process of extraction entails from the provided claim language. These “features” are recited with a high degree of generality without clarification on whether these are simply visual features, or more particularly machine-learned features. The specification generally recites feature extraction [¶0054], that can be anatomical or pathological [¶0062], to generate a feature set [¶0066, 89] but does not provide an explanation of what constitutes a feature and what possible methods can be used to extract said features.
Claims 9, 20 & 27 recite “wherein said images are acquired using a mobile device camera”. While the examiner understands in the context of the claims and specification, that these are images are intended to be specific to the target dataset at the inference stage, these images could be construed as the set of eye images of a cohort of subjects input into the machine learning model during the training stage and are therefore indefinite.
Therefore, claims 1, 10 & 19 are rejected for being indefinite under 35 U.S.C. § 112(b). Subsequently, dependent claims 2, 3, 5-7, 9, 11, 12, 14-16, 18, 20, 21, 23, 24 & 27 are rejected due to their dependence on a claim rejected under 35 U.S.C. § 112(b).
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, 5-7, 10-12, 14-16, 18-21, 23, 24 & 27 rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more.
Claim 1 is reproduced below with additional annotations:
A system comprising:
at least one hardware processor; and
a non-transitory computer-readable storage medium having stored thereon program instructions, the program instruction executable by the at least one hardware processor to:
receive, as input, a set of images of eyes of a cohort of subjects, and a plurality of associated datapoints with respect to each of the subjects in the cohort, wherein at least some of said subjects in said cohort are associated with a diagnosis of a specific eye condition,
extract, with respect to each of said subjects, a set of features representing said images and said datapoints associated with said subject,
at a training stage, train a machine learning model on a training dataset comprising:
all of said sets of features, and
labels indicating a diagnosis of an eye condition associated with each of said subjects, and
at an inference stage, apply said trained machine learning model to a target dataset comprising one or more images and associated datapoints with respect to a target subject, to output a diagnosis of an eye condition in said target subject.
Analysis for claim 1 for subject matter eligibility under the Alice/Mayo framework is as follows:
Step 1: Evaluating whether the claim belongs to one of the statutory categories.
Claim 1 recites a plurality of actions performed by a system, the system being a physical device which is directed to a machine, which falls under one of the statutory categories of invention. (Step 1: Yes)
Step 2A – Prong One: Evaluating whether the claim recites a judicial exception (an abstract enumerated in the 2019 PEG, a law of nature, or a natural phenomenon). If no exception is recited, the claim is eligible. This concludes the eligibility analysis. If the claim recites an exception, go to Step 2A – Prong Two.
Independent claim 1 recites a plurality of mental processes, which fall under the category of abstract ideas. Steps b-e of claim 1, under the broadest reasonable interpretation, recite processes that could be practically performed in the human mind. (The courts consider a mental process (thinking) that “can be performed in the human mind, or by a human using pen and paper” to be an abstract idea.
Step b describes simply receiving a set eye images and associated datapoints for subjects of each eye image. “Receiving” or otherwise obtaining data (i.e., data gathering) is a process wholly capable of being performed by the human mind and simply amounts to insignificant pre-solution activity.
Step c describes extracting features representing said images and datapoints associated with these images. These “features” are recited with a high degree of generality. A human (such as an ophthalmologist) can readily identify and take note of visual features in an image that could be indicative of some eye condition.
Step d describes training a machine learning model on a dataset comprising the aforementioned features and associated labels indicating a diagnosis. This process is recited broadly without any improvement in the neural network itself. An ophthalmologist, by nature of their training, would naturally learn what visual features would be associated with a respective eye condition.
Step e
describes applying the machine learning model to output a diagnosis. Again, this process is recited broadly without any particular improvement in the neural network itself. An ophthalmologist, given their training from previously viewing eye images associated with a particular diagnosis, would readily be able to perform the described inference/diagnosis (i.e., an evaluation or judgement).
These steps would fall into the “mental process” group of abstract ideas – specifically, observations (i.e., data gathering), evaluation, and/or judgement. The limitations, interpreted under their broadest reasonable interpretation consistent with the specification, cover performance of the limitations in the mind or by generic computer components (see MPEP § 2106.04 and the 2019 PEG on subject matter eligibility). (Step 2A – Prong One: Yes).
Step 2A – Prong Two: Evaluating whether the claim recites additional elements that integrate the exception into a practical application of the exception. This evaluation is performed by (a) identifying whether there are any additional elements recited in the claims beyond the judicial exception and (b) evaluating those additional elements individually and in combination to determine whether the claims as a whole integrates the exception into a practical application.
Claim 1 further recites additional elements in the form of (underlined below) –
at least one hardware processor; and
a non-transitory computer-readable storage medium having stored thereon program instructions, the program instruction executable by the at least one hardware processor…
The processor and non-transitory computer-readable storage medium are recited at a high level of generality such that they do not amount to more than generic computer system elements. The claim does not direct to a specific improvement in computers in their communication role or provides a specific improvement in the way computers operate. These limitations generally link the use of the abstract idea to the environment of digital image processing.
The specification descries the invention as seeking to solve problems in telediagnosis by facilitating access to a specialist, however this is not reflected in the claims [¶0004]. Furthermore, the applicant clearly describes human supervision during data gathering of eye images and datapoints as well as and image labeling [¶0070]. The claims however, aren’t directed to solving access to specialist, but merely automating the analysis portion of the process by using a generic machine.
These additional elements neither integrate the abstract idea into a practical application of the abstract idea, nor result in the claim as a whole amounting to significantly more than the abstract idea. (Step 2A – Prong Two: No).
Step 2B: This part of the eligibility analysis evaluates 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.
Claim 1 as a whole does not amount to significantly more than the recited exception, nor does it add an inventive concept to the claim. It merely recites an automated analysis performed by a machine learning model for classification purposes, which is well-understood and routine in the art without substantially more (see MPEP § 2016.04(d) & CAFC – Dental Monitoring SAS vs Align Technology, Inc. 2024-2270). (Step 2B: No).
Therefore, claim 1 is ineligible.
Independent claim 10, which recites a computer-implemented method, is directed to a process, which is a statutory category of invention. Similar analysis from claim 1 is applicable to claim 10 since the recited function is substantially identical.
Claim 10 recites no additional elements. Therefore, claim 10 is ineligible.
Independent claim 19, which recites a computer program product comprising a non-transitory computer-readable storage medium having program instructions, is directed to a process, which is a statutory category of invention. Similar analysis from claim 1 is applicable to claim 10 since the recited function is substantially identical.
Claim 19 recites no additional elements. Therefore, claim 19 is ineligible.
Claims 2, 11 & 20 recite “wherein said specific eye condition is an anterior segment eye condition.” This element merely provides supplemental information about the type of eye conditions input into the machine learning model. They do not integrate the exception into a practical application, nor amount to significantly more than the judicial exception. Claims 2, 11 & 20 are ineligible.
Claims 3, 12 & 21 recite “wherein said images are annotated to indicate anatomical and pathological eye features represented in said images, and wherein said annotations are in the form of one of: an exact outline of each of said anatomical and pathological features, or a bounding box enclosing said each of anatomical and pathological features”. Again, these elements merely further describe annotations on images provided to the machine learning model. They do not integrate the exception into a practical application, nor amount to significantly more than the judicial exception. Claims 3, 12 & 21 are ineligible.
Claims 5, 14 & 23 recite “wherein said datapoints comprise, with respect to each of said subjects, at least one of the following categories of datapoints:
(i) demographic information datapoints;
(ii) medical history datapoints; and
(iii) eye condition signs and symptoms”
These additional elements merely further define the composition of the datapoints outlined in claim 1, and provide little more than simple clarification on the input data used by the machine learning model. They do not integrate the exception into a practical application, nor amount to significantly more than the judicial exception. Claims 5, 14, & 23 are ineligible.
Claims 6, 15 & 24 recite “wherein said labels represent, with respect to each of said subjects, binary values indicating the presence or absence of an anterior eye segment condition, and wherein said diagnosis is expressed as a binary value indicating the presence or absence of an anterior segment eye condition in said target subject”. These additional elements merely elaborate on what the provided labels used in training are. They fail to integrate the exception into a practical application, nor amount to significantly more than the judicial exception. Claims 6 & 15 are ineligible. Claim 24 further recites additional elements described below.
Claims 7, 16 & 24 recite “wherein said labels represent, with respect to each of said subjects, values on a scale indicating a severity level associated with an anterior segment eye condition, and wherein said diagnosis is expressed as a value on a scale indicating a severity level associated with an anterior segment eye condition in said target subject”. This additional element again elaborates on what the provided labels in training are, and fails to integrate the exception into a practical application, nor amount to significantly more than the judicial exception. Claims 7, 16 & 24 are ineligible.
Claim 9, 20 & 27 recite “wherein said images are acquired using a mobile device camera”. This element merely describes that images obtained for subsequent diagnosis are taken on a mobile camera, and just defines a means for image acquisition. This fails to integrate the exception into a practical application, nor amounts to significantly more than the judicial exception. Claims 9, 20 & 27 are ineligible.
Claim Rejections - 35 USC § 102
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claim(s) 1, 5, 9, 10, 14, 18, 19, 23 & 27 is/are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Tran; Bao (US 2020/0405148 A1), hereinafter referred to as "Tran".
Regarding claim 1, Tran teach A system (deep learning systems for eye structure recognitions that may utilize a variety of different types of neural networks [¶0028-40; Figs 2A-2J]) comprising:
at least one hardware processor (the instructions can be executed by a processor of a smartphone [¶0102]); and
a non-transitory computer-readable storage medium having stored thereon program instructions (the non-transitory computer-readable storage medium of a smartphone [¶0102] – the examiner notes that a smartphone inherently requires some type memory or storage), the program instructions executable by the at least one hardware processor to:
receive, as input, a set of images of eyes of a cohort of subjects, and a plurality of associated datapoints with respect to each of the subjects in the cohort, wherein at least some of said subjects in said cohort are associated with a diagnosis of a specific eye condition (an image dataset of patient eye images is obtained [¶0019; Fig. 1A], wherein the dataset includes associated information (i.e., clinical information, which is being interpreted as the associated datapoints) regarding diagnosis and associated symptoms for each patient [¶0097]),
extract, with respect to each of said subjects, a set of features representing said images and said datapoints associated with said subject (features are extracted from the image dataset to be provided to the deep learning network [¶0019, 22-25; Figs 1A-B], clinical information (i.e., associated datapoints such as age, sex, medical history, etc.) is provided during feature extraction [¶0026-27; Fig. 1C]),
at a training stage, train a machine learning model on a training dataset comprising:
(i) all of said sets of features (features from current and previous eye exams are fed into the input layer of a neural network for training [¶0026; Fig. 1C]), and
(ii) labels indicating a diagnosis of an eye condition associated with each of said subjects (clinical information (being interpreted as a form of label) can include information regarding a diagnosis for a patient which inherently act as a label for an eye condition for a given eye image of a subject [¶0026 & 97; Fig. 1C]), and
at an inference stage, apply said trained machine learning model to a target dataset comprising one or more images and associated datapoints with respect to a target subject, to output a diagnosis of an eye condition in said target subject (the system utilizes current and past exam information and fundus images of a patient’s eye stored in an image database to make a prediction of a diagnosis or prognosis [¶0025-27; Figs. 1B-1C]).
With respect to claim 5, Tran teach The system of claim 1 (described above), wherein said datapoints comprise, with respect to each of said subjects, at least one of the following categories of datapoints (the examiner notes that, due to the use of “at least one of … and” when reciting a list of limitations, all limitations within the list must be mapped (see MPEP 2111.01(II) and Superguide Corp. v. DirecTV Enterprises, Inc., 358 F.3d 870, 875, 69 USPQ2d 1865, 1868 (Fed. Cir. 2004)):
(i) demographic information datapoints (clinical information (i.e., the datapoints) may be a patient’s sex or age [¶0026]);
(ii) medical history datapoints (clinical information (i.e., the datapoints) may be a patient’s medical history [¶0026]; and
(iii) eye condition signs and symptoms datapoints (clinical information (i.e., the datapoints) may be diagnosis information and associated symptoms [¶0097]).
As for claim 9, Tran teach The system of claim 1 (described previously), wherein said images are acquired using a mobile device camera (images are acquired via image capture device 1 is a smart-phone 110 [¶0043; Figs. 3A & B]).
Regarding claim 10, Tran teach A computer-implemented method (the high level process outlined in Fig. 1A [¶0019]) comprising:
receiving, as input, a set of images of eyes of a cohort of subjects, and a plurality of associated data points with respect to each of the subjects in the cohort, wherein at least some of said subjects in said cohort are associated with a diagnosis of a specific eye condition (an image dataset of patient eye images is obtained [¶0019; Fig. 1A], wherein the dataset includes associated information (i.e., clinical information, which is being interpreted as the associated datapoints) regarding diagnosis and associated symptoms for each patient [¶0097]);
extracting, with respect to each of said subjects, a set of features representing said images and said datapoints associated with said subject (features are extracted from the image dataset to be provided to the deep learning network [¶0019, 22-25; Figs 1A-B], clinical information (i.e., associated datapoints such as age, sex, medical history, etc.) is provided during feature extraction [¶0026-27; Fig. 1C]);
at a training stage, training a machine learning model on a training dataset comprising:
(i) all of said sets of features (features from current and previous eye exams are fed into the input layer of a neural network for training [¶0026; Fig. 1C]), and
(ii) labels indicating a diagnosis of an eye condition associated with each of said subjects (clinical information (being interpreted as a form of label) can include information regarding a diagnosis for a patient which inherently act as a label for an eye condition for a given eye image of a subject [¶0026 & 97; Fig. 1C]); and
at an inference stage, applying said trained machine learning model to a target dataset comprising one or more images and associated datapoints with respect to a target subject, to output a diagnosis of an eye condition in said target subject (the system utilizes current and past exam information and fundus images of a patient’s eye stored in an image database to make a prediction of a diagnosis or prognosis [¶0025-27; Figs. 1B-1C]).
Concerning claim 14, Tran teach The computer-implemented method of claim 10 (described above), wherein said datapoints comprise, with respect to each of said subjects, at least one of the following categories of datapoints (the examiner notes that, due to the use of “at least one of … and” when reciting a list of limitations, all limitations within the list must be mapped (see MPEP 2111.01(II) and Superguide Corp. v. DirecTV Enterprises, Inc., 358 F.3d 870, 875, 69 USPQ2d 1865, 1868 (Fed. Cir. 2004)):
(i) demographic information datapoints (clinical information (i.e., the datapoints) may be a patient’s sex or age [¶0026]);
(ii) medical history datapoints (clinical information (i.e., the datapoints) may be a patient’s medical history [¶0026]; and
(iii) eye condition signs and symptoms datapoints (clinical information (i.e., the datapoints) may be diagnosis information and associated symptoms [¶0097]).
Considering claim 18, Tran teach The computer-implemented method of claim 10 (described previously), wherein said images are acquired using a mobile device camera (images are acquired via image capture device 1 is a smart-phone 110 [¶0043; Figs. 3A & B]).
Regarding claim 19, A computer program product (the trained neural network is implemented through a smart phone [¶0019]) comprising a non-transitory computer- readable storage medium (the non-transitory computer-readable storage medium of a smartphone [¶0102] – the examiner notes that a smartphone inherently requires some type memory or storage) having program instructions embodied therewith, the program instructions executable by at least one hardware processor (the instructions can be executed by a processor of a smartphone [¶0102]) to:
receive, as input, a set of images of eyes of a cohort of subjects, and a plurality of associated datapoints with respect to each of the subjects in the cohort, wherein at least some of said subjects in said cohort are associated with a diagnosis of a specific eye condition (an image dataset of patient eye images is obtained [¶0019; Fig. 1A], wherein the dataset includes associated information (i.e., clinical information, which is being interpreted as the associated datapoints) regarding diagnosis and associated symptoms for each patient [¶0097]),
extract, with respect to each of said subjects, a set of features representing said images and said datapoints associated with said subject (features are extracted from the image dataset to be provided to the deep learning network [¶0019, 22-25; Figs 1A-B], clinical information (i.e., associated datapoints such as age, sex, medical history, etc.) is provided during feature extraction [¶0026-27; Fig. 1C]),
at a training stage, train a machine learning model on a training dataset comprising:
(i) all of said sets of features (features from current and previous eye exams are fed into the input layer of a neural network for training [¶0026; Fig. 1C]), and
(ii) labels indicating a diagnosis of an eye condition associated with each of said subjects (clinical information (being interpreted as a form of label) can include information regarding a diagnosis for a patient which inherently act as a label for an eye condition for a given eye image of a subject [¶0026 & 97; Fig. 1C]), and
at an inference stage, apply said trained machine learning model to a target dataset comprising one or more images and associated datapoints with respect to a target subject, to output a diagnosis of an eye condition in said target subject (the system utilizes current and past exam information and fundus images of a patient’s eye stored in an image database to make a prediction of a diagnosis or prognosis [¶0025-27; Figs. 1B-1C]).
With respect to claim 23, Tran teach The computer program product of claim 19 (as described above), wherein said datapoints comprise, with respect to each of said subjects, at least one of the following categories of datapoints (the examiner notes that, due to the use of “at least one of … and” when reciting a list of limitations, all limitations within the list must be mapped (see MPEP 2111.01(II) and Superguide Corp. v. DirecTV Enterprises, Inc., 358 F.3d 870, 875, 69 USPQ2d 1865, 1868 (Fed. Cir. 2004)):
(i) demographic information datapoints (clinical information (i.e., the datapoints) may be a patient’s sex or age [¶0026]);
(ii) medical history datapoints (clinical information (i.e., the datapoints) may be a patient’s medical history [¶0026]; and
(iii) eye condition signs and symptoms datapoints (clinical information (i.e., the datapoints) may be diagnosis information and associated symptoms [¶0097]).
Turning to claim 27, Tran teach The computer program product of claim 19 (described previously), wherein said images are acquired using a mobile device camera (images are acquired via image capture device 1 is a smart-phone 110 [¶0043; Figs. 3A & B]).
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claim(s) 2, 3, 11, 12, 20, 21 is/are rejected under 35 U.S.C. 103 as being unpatentable over Tran; Bao (US 2020/0405148 A1), hereinafter referred to as "Tran" in view of Xu et al (“Deep Sequential Feature Learning in Clinical Image Classification of Infectious Keratitis”, 2020, Engineering), hereinafter referred to as “Xu”.
Regarding claim 2, Tran teach The system of claim 1 (described previously), and while Tran generally discloses generally discloses inspection of anterior segments of the eye to photograph cataracts, they fail to explicitly disclose using these as training data.
Xu, however, is analogous art pertinent to the field of endeavor of the present application and describe utilizing a sequential deep learning model to efficiently classy clinical images of infection corneal diseases. More particularly, Xu teach wherein said specific eye condition is an anterior segment eye condition (Xu: a dataset of 2284 images of 867 patients with varying types of keratitis [Sec 2A: Image Datasets - 02-03; Fig. 1] wherein keratitis is a disease of the cornea [Sec: Abstract] - the examiner notes that the cornea is component of the anterior segment of the eye).
Xu state that timing for the rapid diagnosis of infectious keratitis is critical for prompt and precise treatment given the speed of its progression. They explain that their model provides a facile and accurate identification of infectious keratitis [Sec: Abstract]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to utilize the training data provided by Xu to better inform the diagnosis platform of Tran for identifying anterior segment eye conditions such as keratitis.
Turning to claim 3, Tran teach The system of claim 1 (described previously), wherein said images are annotated to indicate anatomical and
exact outline of each of said anatomical (Tran: anatomical eye features (in the form of blood vessels) are segmented to provide an outline of the vessels for each eye (illustrated in Fig. 1B), wherein each pixel can be assigned a label for vessel vs non-vessel classification [¶0021-25; Fig. 1A]), or a bounding box enclosing said each of anatomical (the examiner notes that, given the use of the disjunctive "or", only one of the listed limitations requires mapping).
While Tran implies diagnosis via extraction of features related to various eye pathologies, they fail to describe obtaining an exact outline or bounding box of pathological features. Xu, on the other hand, teach exact outline of each of said (Xu: lesions of the cornea are annotated via outline of the keratits lesion area of the cornea [Sec 2B: Sequential-level learning based diagnostic deep models - ¶03-5; Fig. 2]), or a bounding box enclosing said each of (the examiner notes that, given the use of the disjunctive "or", only one of the listed limitations requires mapping).
Xu state that timing for the rapid diagnosis of infectious keratitis is critical for prompt and precise treatment given the speed of its progression. They explain that their model provides a facile and accurate identification of infectious keratitis [Sec: Abstract]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to utilize the training data provided by Xu to better inform the diagnosis platform of Tran for identifying anterior segment eye conditions such as keratitis.
With respect to claim 11, Tran teach The computer-implemented method of claim 10 (described previously), and while Tran generally discloses generally discloses inspection of anterior segments of the eye to photograph cataracts, they fail to explicitly disclose using these as training data.
More particularly, Xu teach wherein said specific eye condition is an anterior segment eye condition (Xu: a dataset of 2284 images of 867 patients with varying types of keratitis [Sec 2A: Image Datasets - 02-03; Fig. 1] wherein keratitis is a disease of the cornea [Sec: Abstract] - the examiner notes that the cornea is component of the anterior segment of the eye).
Xu state that timing for the rapid diagnosis of infectious keratitis is critical for prompt and precise treatment given the speed of its progression. They explain that their model provides a facile and accurate identification of infectious keratitis [Sec: Abstract]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to utilize the training data provided by Xu to better inform the diagnosis platform of Tran for identifying anterior segment eye conditions such as keratitis.
As for claim 12, Tran teach The computer-implemented method of claim 10 (described previously), wherein said images are annotated to indicate anatomical and
exact outline of each of said anatomical (Tran: anatomical eye features (in the form of blood vessels) are segmented to provide an outline of the vessels for each eye (illustrated in Fig. 1B), wherein each pixel can be assigned a label for vessel vs non-vessel classification [¶0021-25; Fig. 1A]), or a bounding box enclosing said each of anatomical (the examiner notes that, given the use of the disjunctive "or", only one of the listed limitations requires mapping).
While Tran implies diagnosis via extraction of features related to various eye pathologies, they fail to describe obtaining an exact outline or bounding box of pathological features. Xu, on the other hand, teach exact outline of each of said (Xu: lesions of the cornea are annotated via outline of the keratits lesion area of the cornea [Sec 2B: Sequential-level learning based diagnostic deep models - ¶03-5; Fig. 2]), or a bounding box enclosing said each of (the examiner notes that, given the use of the disjunctive "or", only one of the listed limitations requires mapping).
Xu state that timing for the rapid diagnosis of infectious keratitis is critical for prompt and precise treatment given the speed of its progression. They explain that their model provides a facile and accurate identification of infectious keratitis [Sec: Abstract]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to utilize the training data provided by Xu to better inform the diagnosis platform of Tran for identifying anterior segment eye conditions such as keratitis.
Concerning claim 20, Tran teach The computer program product of claim 19 (described previously), and while Tran generally discloses generally discloses inspection of anterior segments of the eye to photograph cataracts, they fail to explicitly disclose using these as training data.
More particularly, Xu teach wherein said specific eye condition is an anterior segment eye condition (Xu: a dataset of 2284 images of 867 patients with varying types of keratitis [Sec 2A: Image Datasets - 02-03; Fig. 1] wherein keratitis is a disease of the cornea [Sec: Abstract] - the examiner notes that the cornea is component of the anterior segment of the eye).
Xu state that timing for the rapid diagnosis of infectious keratitis is critical for prompt and precise treatment given the speed of its progression. They explain that their model provides a facile and accurate identification of infectious keratitis [Sec: Abstract]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to utilize the training data provided by Xu to better inform the diagnosis platform of Tran for identifying anterior segment eye conditions such as keratitis.
With regard to claim 21, Tran teach The computer program product of claim 19 (described previously), wherein said images are annotated to indicate anatomical and
exact outline of each of said anatomical (Tran: anatomical eye features (in the form of blood vessels) are segmented to provide an outline of the vessels for each eye (illustrated in Fig. 1B), wherein each pixel can be assigned a label for vessel vs non-vessel classification [¶0021-25; Fig. 1A]), or a bounding box enclosing said each of anatomical (the examiner notes that, given the use of the disjunctive "or", only one of the listed limitations requires mapping).
While Tran implies diagnosis via extraction of features related to various eye pathologies, they fail to describe obtaining an exact outline or bounding box of pathological features. Xu, on the other hand, teach exact outline of each of said (Xu: lesions of the cornea are annotated via outline of the keratits lesion area of the cornea [Sec 2B: Sequential-level learning based diagnostic deep models - ¶03-5; Fig. 2]), or a bounding box enclosing said each of (the examiner notes that, given the use of the disjunctive "or", only one of the listed limitations requires mapping).
Xu state that timing for the rapid diagnosis of infectious keratitis is critical for prompt and precise treatment given the speed of its progression. They explain that their model provides a facile and accurate identification of infectious keratitis [Sec: Abstract]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to utilize the training data provided by Xu to better inform the diagnosis platform of Tran for identifying anterior segment eye conditions such as keratitis.
Claim(s) 6, 15 & 24 is/are rejected under 35 U.S.C. 103 as being unpatentable over Tran; Bao (US 2020/0405148 A1), hereinafter referred to as "Tran" in view of Zhang et al ("Attention-Based Multi-Model Ensemble for Automatic Cataract Detection in B-Scan Eye Ultrasounds", IJCNN, 2020), hereinafter referred to as “Zhang”.
Regarding claim 6, Tran teach The system of claim 1 (as described previously), and while Xu describe each image having clinical information conveying a diagnosis of eye images for training, they fail to describe a binary classification for whether these diagnoses are of an anterior segment eye condition.
Zhang, per contra, is analogous art pertinent to the field of endeavor and disclose an attention-based neural network for cataract detection. Zhang teach wherein said labels represent, with respect to each of said subjects, binary values indicating the presence or absence of an anterior segment eye condition, and wherein said diagnosis is expressed as a binary value indicating the presence or absence of an anterior segment eye condition in said target subject (Zhang: B-scan eye ultrasound images of either normal or cataract-afflicted eyes are used for training, wherein normal eyes are labeled as "0" while eyes with cataracts are labeled with "1" [Sec IV-A. Dataset and Metrics - ¶01-03; Fig. 6; Table I] - the examiner notes that cataracts are a disorder of the lens of the eye, with the lens being a component of the anterior segment of the eye).
Zhang further elaborates detection of cataracts in the early stage is critical for preventing blindness in patients, with their model detecting cataracts accurately 97.5% of the time [Sec: Abstract] One of ordinary skill in the art before the effective filing date of the present application would recognize the advantage of utilizing the labeled B-scan eye ultrasound images to train the base system of Tran for effectively diagnosing cataracts from patient eye images.
Turning to claim 15, Tran teach The computer implemented method of claim 10 (as described previously), and while Xu describe each image having clinical information conveying a diagnosis of eye images for training, they fail to describe a binary classification for whether these diagnoses are of an anterior segment eye condition.
Zhang, in contrast, teach wherein said labels represent, with respect to each of said subjects, binary values indicating the presence or absence of an anterior segment eye condition, and wherein said diagnosis is expressed as a binary value indicating the presence or absence of an anterior segment eye condition in said target subject (Zhang: B-scan eye ultrasound images of either normal or cataract-afflicted eyes are used for training, wherein normal eyes are labeled as "0" while eyes with cataracts are labeled with "1" [Sec IV-A. Dataset and Metrics - ¶01-03; Fig. 6; Table I] - the examiner notes that cataracts are a disorder of the lens of the eye, with the lens being a component of the anterior segment of the eye).
Zhang further elaborates detection of cataracts in the early stage is critical for preventing blindness in patients, with their model detecting cataracts accurately 97.5% of the time [Sec: Abstract] One of ordinary skill in the art before the effective filing date of the present application would recognize the advantage of utilizing the labeled B-scan eye ultrasound images to train the base system of Tran for effectively diagnosing cataracts from patient eye images.
Concerning claim 24, Tran teach The computer program product of claim 19 (as previously described), but they fail to describe the labels for the training images comprising a binary classification for the presence of a condition of the anterior eye segment, nor do they recite a grading of the severity.
Zhang, however, teach wherein said labels represent, with respect to each of said subjects, binary values indicating the presence or absence of an anterior segment eye condition, and wherein said diagnosis is expressed as a binary value indicating the presence or absence of an anterior segment eye condition in said target subject (Zhang: B-scan eye ultrasound images of either normal or cataract-afflicted eyes are used for training, wherein normal eyes are labeled as "0" while eyes with cataracts are labeled with "1" [Sec IV-A. Dataset and Metrics - ¶01-03; Fig. 6; Table I] - the examiner notes that cataracts are a disorder of the lens of the eye, with the lens being a component of the anterior segment of the eye), or wherein said labels represent, with respect to each of said subjects, values on a scale indicating a severity level associated with an anterior segment eye condition, and wherein said diagnosis is expressed as a value on a scale indicating a severity level associated with an anterior segment eve condition in said target subject (given the use of the disjunctive “or” delineating what either of said labels can be represented as, only one of these limitations needs to be mapped to).
Zhang further elaborates detection of cataracts in the early stage is critical for preventing blindness in patients, with their model detecting cataracts accurately 97.5% of the time [Sec: Abstract] One of ordinary skill in the art before the effective filing date of the present application would recognize the advantage of utilizing the labeled B-scan eye ultrasound images to train the base system of Tran for effectively diagnosing cataracts from patient eye images.
Claim(s) 7 & 16 is/are rejected under 35 U.S.C. 103 as being unpatentable over Tran; Bao (US 2020/0405148 A1), hereinafter referred to as "Tran" in view of Gao et al ("Automatic Feature Learning to Grade Nuclear Cataracts Based on Deep Learning”, IEEE, 2015), hereinafter referred to as “Gao”.
Regarding claim 7, Tran teach The system of claim 1 (as previously described), however, Tran is silent with respect to training data indicating a severity level associated with an anterior segment eye condition.
Gao, on the other hand, describe utilizing ophthalmologist-graded images for training a machine learning model to recognize and grade images of patient’s cataracts. Gao further teach wherein said labels represent, with respect to each of said subjects, values on a scale indicating a severity level associated with an anterior segment eye condition, and wherein said diagnosis is expressed as a value on a scale indicating a severity level associated with an anterior segment eye condition in said target subject (Gao: the ACHIKO-NC dataset is used for training the neural network, consisting of 5378 images of cataracts labeled with a grading score of 0.1-5.0, with a higher decimal score indicating a greater severity [Sec A. Database - ¶01-02; Table I] - the examiner again notes that cataracts are a disorder of the lens of the eye, with the lens being a component of the anterior segment of the eye.).
Gao further clarify that assessing the severity of cataracts is essential for monitoring progression, however current grading methods may not take into account all image features that may be indicative of a cataract. They address this by proposing a method that automatically learns features for grading cataract severity [Sec: Abstract]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to utilize the graded ACHIKO-NC dataset to train the base system of Tran to not only detect but also grade instances of cataracts in eye images.
Turning to claim 16, Tran teach The computer-implemented method of claim 1 (as previously described), however, Tran is silent with respect to training data indicating a severity level associated with an anterior segment eye condition.
Gao, per contra, teach wherein said labels represent, with respect to each of said subjects, values on a scale indicating a severity level associated with an anterior segment eye condition, and wherein said diagnosis is expressed as a value on a scale indicating a severity level associated with an anterior segment eye condition in said target subject (Gao: the ACHIKO-NC dataset is used for training the neural network, consisting of 5378 images of cataracts labeled with a grading score of 0.1-5.0, with a higher decimal score indicating a greater severity [Sec A. Database - ¶01-02; Table I] - the examiner again notes that cataracts are a disorder of the lens of the eye, with the lens being a component of the anterior segment of the eye.).
Gao further clarify that assessing the severity of cataracts is essential for monitoring progression, however current grading methods may not take into account all image features that may be indicative of a cataract. They address this by proposing a method that automatically learns features for grading cataract severity [Sec: Abstract]. It would have been obvious to one of ordinary skill in the art before the effective filing date of the present application to utilize the graded ACHIKO-NC dataset to train the base system of Tran to not only detect but also grade instances of cataracts in eye images.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Nakazawa et al (US 2023/0337908 A1) teach an ophthalmic feature extractor and classifier to estimate a pathological condition of a subject’s eyes.
Zhang et al (US 2019/0110753 A1) describe utilizing a deep learning algorithm implemented through a mobile device to analyze and diagnose ophthalmic diseases.
Tong et al (“Application of machine learning in ophthalmic imaging modalities, 2020, Eye and Vision) provide a review of the state of the art for machine learning-based classification of ophthalmic conditions.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Michael M. Sofroniou whose telephone number is (571)272-0287. The examiner can normally be reached M-F: 8:30 AM - 5:00 PM.
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, John M. Villecco can be reached at (571) 272-7319. 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.
/MICHAEL M SOFRONIOU/Examiner, Art Unit 2661
/AARON W CARTER/Primary Examiner, Art Unit 2661
Prosecution Insights