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 .
Priority
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55. Acknowledgement is made of applicant’s claim for foreign priority to 17 May 2024 under 35 U.S.C. 119(a)-(d).
Status of Claims
Claims 1-14 received on 16 May 2025 are currently pending and being considered by Examiner in this Office 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 1-14 are rejected under 35 U.S.C. 101 because the claimed invention is directed to a judicial exception (i.e., a law of nature, a natural phenomenon, or an abstract idea) without significantly more.
The claims recite subject matter within a statutory category as a process (claims 1-7) and machine (claims 8-14) (Subject Matter Eligibility (SME) Test Step 1: Yes) which recite steps of:
generating analysis data from facial photograph information of an analysis subject;
storing response data of an obstructive sleep apnea (OSA) screening questionnaire of the analysis subject in the memory;
inputting the analysis data and the response data into a pre-trained machine learning model and inferring information about the degree of OSA; and
transmitting the inference result to at least one terminal or outputting the inference result to a display.
These steps of generating analysis data from facial photograph information, storing response data of an OSA screening questionnaire, inputting the analysis data and response data into a model and inferring information about the degree of OSA, and transmitting the inference result, as drafted, under the broadest reasonable interpretation, includes performance of the limitation in the mind but for recitation of generic computer components. That is, other than reciting steps as performed by the generic computer components, nothing in the claim element precludes the step from practically being performed in the mind. For example, but for the generating analysis data language, generating analysis data in the context of this claim encompasses a mental process of the user, such as a sleep health specialist determining whether a user may have complications with sleep apnea based on facial shape. Similarly, the limitation of storing response data of an OSA screening questionnaire, as drafted, is a process that, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, such as the doctor receiving varying answers for a patient’s questionnaire regarding OSA when considering/determining whether the patient may have OSA. For example, but for the inputting the analysis data and response data into a model language, inputting the analysis data in the context of this claim encompasses a mental process of the user or doctor utilizing a mathematical or health model to determine whether a patient has OSA based on the varying data collected by the doctor and sending or transmitting a result of the applied model/. While the model is claimed to be a pre-trained machine learning model, it is understood by Examiner that this merely represents efforts of “apply it”. That is, while already-existing, generic components , e.g. pre-trained machine learning models, if a claim limitation, under its broadest reasonable interpretation, covers performance of the limitation in the mind but for the recitation of generic computer components, then it falls within the “Mental Processes” grouping of abstract ideas. Accordingly, the claim recites an abstract idea.
These steps of generating analysis data from facial photograph information, storing response data of an OSA screening questionnaire, inputting the analysis data and response data into a model and inferring information about the degree of OSA, and transmitting the inference result, as drafted, under the broadest reasonable interpretation, as drafted, under the broadest reasonable interpretation, includes methods of organizing human activity. MPEP 2106.04(a)(2)(II) sets forth various methods of organizing human activity, including concepts relating to fundamental economic principles or practices (including hedging, insurance, mitigating risk); commercial or legal interactions (including agreements in the form of contracts, legal obligations, advertising, marketing or sales activities or behaviors, and business relations); and managing personal behavior or relationships or interactions between people, (including social activities, teaching, and following rules or instructions). For instance, the steps recited relate to managing personal behavior or relationships or interactions between people, (including social activities, teaching, and following rules or instructions) at least by the example provided in MPEP 2106.04(a)(2)(II)(C) regarding a mental process that a neurologist should follow when testing a patient for nervous system malfunctions. That is, the steps recited relate heavily to a mental process that can be followed by a diagnosis entity regarding a patient’s OSA. Accordingly, the claim recites an abstract idea.
Dependent claims recite additional subject matter which further narrows or defines the abstract idea embodied in the claims (such as claim 2-7 & 9-14, reciting particular aspects of how may be performed in the mind but for recitation of generic computer components) (SME Test Step 2A, Prong 1: Yes).
This judicial exception is not integrated into a practical application. In particular, the additional elements do not integrate the abstract idea into a practical application, other than the abstract idea per se, because the additional elements amount to no more than limitations which:
amount to mere instructions to apply an exception (such as recitation of a memory, a processor, a terminal, and a display amounts to invoking computers as a tool to perform the abstract idea, see Applicant’s Specification [0124] for a memory; Spec [0099] for a processor; Spec [0057] for a terminal; Spec [0096] for a display, see MPEP 2106.05(f));
add insignificant extra-solution activity to the abstract idea (such as recitation of receiving facial photograph information of an analysis subject, receiving response data of an OSA screening questionnaire of the of the analysis subject, inputting the analysis data and the response data into a pre-trained machine learning model, transmitting the inference result to at least one terminal amounts to mere data gathering; recitation of inputting the analysis data into a pre-trained machine learning model amounts to selecting a particular data source or type of data to be manipulated, recitation of inferring information about the degree of OSA amounts to insignificant application, see MPEP 2106.05(g); recitation of outputting the inference result to a display amounts to gathering and analyzing information using conventional techniques and displaying the result, TLI Communications);
generally link the abstract idea to a particular technological environment or field of use (such as the method/apparatus being used for predicting a degree of obstructive sleep apnea (OSA), see MPEP 2106.05(h)).
Dependent claims recite additional subject matter which amount to limitations consistent with the additional elements in the independent claims (such as claims 2-7 & 9-14, which recite limitations relating to a processor, additional limitations which amount to invoking computers as a tool to perform the abstract idea, see Applicant’s Specification [0099] for a processor, see MPEP 2106.05(f); claims 2-6 & 9-13, which recite limitations relating to inputting the facial photograph inputting the first analysis data and response data, extracting a plurality of landmark information from the facial photograph information, inputting data into a first, second, third, etc., model, extracting varying information, additional limitations which add insignificant extra-solution activity to the abstract idea which amounts to mere data gathering; claims 2, 4, 6, 9, 11, & 13, which recite limitations relating to generating analysis data based on extracting a plurality of landmark information and second analysis data using distance information between the landmarks, and/or generating other information/data based on varying received data, additional limitations which add insignificant extra-solution activity to the abstract idea by selecting a particular data source or type of data to be manipulated; claims 2-7 & 9-14, which recite limitations relating to inferring information about the degree of OSA, or using a machine learning model to infer information, additional limitations which amount to insignificant application; claims 2-7 & 9-14, which generally recite additional limitations which generally link the abstract idea to a particular technological environment or field of use). Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely provide conventional computer implementation and do not impose a meaningful limit to integrate the abstract idea into a practical application (SME Test Step 2A, Prong 2: No).
The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. As discussed above with respect to discussion of integration of the abstract idea into a practical application, the additional elements amount to no more than mere instructions to apply an exception, add insignificant extra-solution activity to the abstract idea, and generally link the abstract idea to a particular technological environment or field of use. Additionally, the additional limitations, other than the abstract idea per se, amount to no more than limitations which:
amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields (such as receiving facial photograph information of an analysis subject, receiving response data of an OSA screening questionnaire of the of the analysis subject, inputting the analysis data and the response data into a pre-trained machine learning model, transmitting the inference result to at least one terminal, e.g., receiving or transmitting data over a network, Symantec, MPEP 2106.05(d)(II)(i); inferring information about the degree of OSA, generating varying data, such as by input into a model, e.g., performing repetitive calculations, Flook, MPEP 2106.05(d)(II)(ii); maintaining one or more questionnaire or analysis data of a patient, such as in an electronic health record, under BRI, e.g., electronic recordkeeping, Alice Corp., MPEP 2106.05(d)(II)(iii); storing received data such as response data of an OSA screening questionnaire or analysis data of a patient, storing computerized instructions for performance of the steps recited throughout the claims, e.g., storing and retrieving information in memory, Versata Dev. Group, MPEP 2106.05(d)(II)(iv); receiving response data of an OSA screening questionnaire, which under BRI, could merely include electronic scanning/extraction from said questionnaire to generate/obtain the response data, e.g., electronic scanning or extracting data from a physical document, Content Extraction, MPEP 2106.05(d)(II)(v); responding to an OSA screening questionnaire, such as at a UI or digital interface, e.g., a web browser’s back and forward button functionality, Internet Patent Corp., MPEP 2106.05(d)(II)(ii); inferring information about the degree of OSA, generating varying data, such as by input into a model, i.e. via machine learning models, e.g., see Gill and Hogg which disclose generally WURC activity of applying machine learning models for determinations of health risks, and more specifically OSA).
Dependent claims recite additional subject matter which, as discussed above with respect to integration of the abstract idea into a practical application, amount to invoking computers as a tool to perform the abstract idea. Dependent claims recite additional subject matter which amount to limitations consistent with the additional elements in the independent claims (such as claims 2-7 & 9-14, additional limitations which amount to elements that have been recognized as well-understood, routine, and conventional activity in particular fields, claims 2-6 & 9-13, which recite limitations relating to inputting the facial photograph inputting the first analysis data and response data, extracting a plurality of landmark information from the facial photograph information, inputting data into a first, second, third, etc., model, extracting varying information, e.g., receiving or transmitting data over a network, Symantec, MPEP 2106.05(d)(II)(i); claims 2, 4, 6, 9, 11, & 13, which recite limitations relating to generating analysis data based on extracting a plurality of landmark information and second analysis data using distance information between the landmarks, and/or generating other information/data based on varying received data, e.g., performing repetitive calculations, Flook, MPEP 2106.05(d)(II)(ii); claims 2, 4, 6, 9, 11, & 13, which recite limitations relating to maintaining one or more questionnaire response data or analysis data of a patient, such as in an electronic health record, under BRI, e.g., electronic recordkeeping, Alice Corp., MPEP 2106.05(d)(II)(iii); claims 2-7 & 9-14, which recite limitations relating to storing received data such as response data of an OSA screening questionnaire or analysis data of a patient, storing computerized instructions for performance of the steps recited throughout the claims, e.g., storing and retrieving information in memory, Versata Dev. Group, MPEP 2106.05(d)(II)(iv); claims 2, 4, 6, 9, 11, & 13, which recite limitations relating to extracting data from one or more questionnaire response data or analysis data of a patient, which includes scanning/extraction from a physical document under BRI, e.g., electronic scanning or extracting data from a physical document, Content Extraction, MPEP 2106.05(d)(II)(v); claims 2, 4, 6, 9, 11, & 13, which recite limitations relating to generating analysis data based on extracting a plurality of landmark information and second analysis data using distance information between the landmarks, and/or generating other information/data based on varying received data,, such as via machine learning models, e.g., see Gill and Hogg which disclose WURC activity for applying machine learning models for determinations of health risks, and more specifically OSA). Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely provide conventional computer implementation (SME Test Step 2B: No).
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or
nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-14 are rejected under 35 U.S.C. 103 as being unpatentable over Lin et al. (“Facial Image Classification for Obstructive Sleep Apnea Pre-Screening” – Reference U - NPL – 2021”), hereinafter Lin, in view of Gill et al. (U.S. Patent Publication No. 2024/0046461), hereinafter “Gill”.
Claim 1 –
Regarding Claim 1, Lin discloses a method for predicting a degree of obstructive sleep apnea (OSA), the method comprising:
generating analysis data from facial photograph information of an analysis subject (See Lin Box 2 which discloses the paper proposing a Resnet net-based model for pre-screening of OSA diagnosis based on 2D face recognition of 2D facial photos);
inputting the analysis data and the response data into a pre-trained machine learning model (See Lin Box 2 which discloses the paper proposing a Resnet net-based model for pre-screening of OSA diagnosis; See Lin Box 3 which discloses a machine learning algorithm being used to analyze the human face and Lin Box 11 which states that the analysis is subsequently being fed into a neural network for extraction and classification, such that the analysis data; See Lin Box 8 which discloses the training data used to produce the pre-trained machine learning model that is used for producing the OSA diagnosis) and inferring information about the degree of OSA (See Lin Box 1 which discloses the paper proposing a Resnet net-based model for pre-screening of OSA diagnosis; See Lin Box 9 and Lin Box 10 which disclose the outputted data including degree of said OSA diagnosis, such as mild, moderate, and severe); and
transmitting the inference result to at least one terminal or outputting the inference result to a display (See Lin Box 10 which discloses displaying the decision result of the model in the form of saliency map, albeit not “transmitted” or a “terminal”, however this limitation has an “or” which means only one of these circumstances has to be met under BRI).
While Lin generally discloses predicting OSA based on facial recognition/analysis of various structures on the patient’s face and it is generally understood that the machine learning algorithm and/or neural network implementations would more than likely include a processing component and storage component for processing stored computerized instructions, Lin is generally silent on the following limitations and for the sake of advancing prosecution, an additional reference will be applied hereinafter:
a method for predicting a degree of obstructive sleep apnea by executing at least one instruction stored in a memory by a processor;
storing response data of an OSA screening questionnaire of the analysis subject in the memory.
However, Gill discloses a method for predicting a degree of obstructive sleep apnea by executing at least one instruction stored in a memory by a processor (See Gill Par [0041]-[0042] which discloses the processes recited throughout Gill being performed via an application, which may be stored on a computer-readable medium, such as memory/data storage, includes programmed instructions for processor to perform said processes/tasks) and storing response data of an OSA screening questionnaire of the analysis subject in the memory (See Gill Par [0041]-[0042] which discloses the processes recited throughout Gill being performed via an application, which may be stored on a computer-readable medium, such as memory/data storage, includes programmed instructions for processor to perform said processes/tasks; See Gill Par [0103] which discloses scan data may be obtained from a storage device, i.e. stored in said storage device in order for retrieval to occur; See Gill Par [0089]-[0091] which discloses collecting subjective data via questionnaire with questions to gather data on symptoms of sleep disorders such that the collected patient input data may be assigned to a patient database, i.e. in storage/memory). The disclosure of Gill is directly applicable to the disclosure of Lin, because the disclosures share limitations and capabilities, namely they are both directed towards determining one or more disorders based on patient data, such as obstructive sleep apnea.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the disclosure of Lin, which already discloses predicting OSA based on facial recognition/analysis of various structures on the patient’s face (and most likely includes processing and storage efforts such as via a computer), to further specifically include processing efforts by a processor and computer storage of the determined data/analysis performed, such as in memory, as disclosed by Gill, because this allows for implementation of the methods in the form of an application that can be executed by a computer, mobile, phone, and/or tablet device (See Gill Par [0041]-[0042]).
Claim 2 –
Regarding Claim 2, Lin and Gill discloses the method of claim 1 in its entirety. Lin further discloses a method, wherein:
the generating of the analysis data includes inputting the facial photograph information into a first machine learning model (See Lin Box 1 which discloses the paper proposing a Resnet net-based model, i.e. learning model, for pre-screening of OSA diagnosis based on 2D face recognition of 2D facial photos), and
generating first analysis data inferring the degree of OSA by the first machine learning model (See Lin Box 1 which discloses the paper proposing a Resnet net-based model, i.e. learning model, for pre-screening of OSA diagnosis; See Lin Box 9 and Box 10 which disclose the outputted data including degree of said OSA diagnosis, such as mild, moderate, and severe).
Claim 3 –
Regarding Claim 3, Lin and Gill disclose the method of claim 2 in its entirety. Lin and Gill further disclose a method, wherein:
the inferring of the information about the degree of OSA includes inputting the first analysis data and the response data into a second machine learning model (See Lin Box 7 which discloses utilizing varying learning models after feature extraction to classify faces based on DCNN and for the diagnosis of OSA; See Lin Box 1 which discloses the paper proposing a Resnet net-based model for pre-screening of OSA diagnosis; See Lin Box 3 which discloses a machine learning algorithm being used to analyze the human face and Box 11 which states that the analysis is subsequently being fed into a neural network for extraction and classification, such that the analysis data; See Lin Box 8 which discloses the training data used to produce the pre-trained machine learning model that is used for producing the OSA diagnosis; it is understood by Examiner that optimizing or determining the amount of learning models to be applied amounts to mere optimization within prior art conditions or through routine experimentation (See MPEP 2144.05(II)) such that specifying the amount of machine learning models beyond two models (i.e. specifying that the model is an ensemble or cascaded machine learning model) is understood to not necessarily hold patentable weight beyond specifying that the model is an ensemble model or comprising more than one learning model in tandem; therefore, see Gill Par [0100] which discloses the potential learning models comprising one or more models, such as in a decision tree ensemble, thereby constituting more than one learning model in tandem; See Lin Box 1 which discloses the paper proposing a Resnet net-based model for pre-screening of OSA diagnosis; See Lin Box 9 & Box 10 which disclose the outputted data including degree of said OSA diagnosis, such as mild, moderate, and severe), and
inferring the information about the degree of OSA by the second machine learning model and generating the inference result (See Lin Box 7 which discloses utilizing varying learning models after feature extraction to classify faces based on DCNN and for the diagnosis of OSA; See Lin “Introduction” which discloses the paper proposing a Resnet net-based model for pre-screening of OSA diagnosis; See Lin Box 3 which discloses a machine learning algorithm being used to analyze the human face and Lin Box 1 which states that the analysis is subsequently being fed into a neural network for extraction and classification, such that the analysis data; See Lin Box 8 which discloses the training data used to produce the pre-trained machine learning model that is used for producing the OSA diagnosis; it is understood by Examiner that optimizing or determining the amount of learning models to be applied amounts to mere optimization within prior art conditions or through routine experimentation (See MPEP 2144.05(II)) such that specifying the amount of machine learning models beyond two models (i.e. specifying that the model is an ensemble or cascaded machine learning model) is understood to not necessarily hold patentable weight beyond specifying that the model is an ensemble model or comprising more than one learning model in tandem; therefore, see Gill Par [0100] which discloses the potential learning models comprising one or more models, such as in a decision tree ensemble, thereby constituting more than one learning model in tandem).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the already-combined disclosure of Lin and Gill with further teachings from Gill regarding the use of multiple machine learning models in tandem or cascaded models, because this allows for multiple nodes or algorithms and adjusting internal derived calculated weights between each of the established node connections by minimizing an error function against actual values during the training process at each node and/or subsequent node (See Gill Par [0100]).
Claim 4 –
Regarding Claim 4, Lin and Gill disclose the method of claim 1 in its entirety. Lin further discloses a method, wherein:
generating of the analysis data includes extracting a plurality of landmark information from the facial photograph information (See Lin Box 1 which discloses that patients may have specific face shapes or structures that affect airway space, leading to an increased risk of OSA, such that specific screening of craniofacial structure, i.e. landmarks, can be considered as an alternative to typical upper airway imaging; See Lin Box 4 which discloses analyzing and comparing some features of craniofacial structure, such that there is a relationship between OSA and some features of craniofacial structure, such as the obvious linear relationship between OSA, neck circumference, mandibular width, and maxillary volume; See Lin Box 8 which discloses in the diagnosis of OSA, focusing on the data of craniofacial structures, i.e. landmark information under BRI, including mandible, temporomandibular, and lateral faces; See Lin Box 5 which discloses identifying, i.e. extracting face key points and then using existing face feature point library to extract and detect them), and
generating second analysis data, which is distance information between the landmarks, by using the plurality of landmark information (See Lin Box 5 which discloses identifying, i.e. extracting face key points and then using existing face feature point library to extract and detect them; See Lin Box 6 which discloses feature images being extracted for analysis, such that geometric features and texture features, can be used to specify characteristics, such as cervicomental contour area, face/mandibular width (i.e. distance under BRI), and tragion-ramus-stomion angle).
Claim 5 –
Regarding Claim 5, Lin and Gill discloses the method of claim 4 in its entirety. Lin and Gill further disclose a method, wherein:
inferring of the information about the degree of OSA includes inputting the second analysis data and the response data into a third machine learning model (It is understood by Examiner that optimizing or determining the amount of learning models to be applied amounts to mere optimization within prior art conditions or through routine experimentation (See MPEP 2144.05(II)) such that specifying the amount of machine learning models beyond two models (i.e. specifying that the model is an ensemble or cascaded machine learning model) is understood to not necessarily hold patentable weight beyond specifying that the model is an ensemble model or comprising more than one learning model in tandem; therefore, see Gill Par [0100] which discloses the potential learning models comprising one or more models, such as in a decision tree ensemble, thereby constituting more than one learning model in tandem; See Lin Box 1 which discloses the paper proposing a Resnet net-based model for pre-screening of OSA diagnosis; See Lin Box 9 & Box 10 which disclose the outputted data including degree of said OSA diagnosis, such as mild, moderate, and severe), and
inferring the information about the degree of OSA by the third machine learning model and generating the inference result (It is understood by Examiner that optimizing or determining the amount of learning models to be applied amounts to mere optimization within prior art conditions or through routine experimentation (See MPEP 2144.05(II)) such that specifying the amount of machine learning models beyond two models (i.e. specifying that the model is an ensemble or cascaded machine learning model) is understood to not necessarily hold patentable weight beyond specifying that the model is an ensemble model or comprising more than one learning model in tandem; therefore, see Gill Par [0100] which discloses the potential learning models comprising one or more models, such as in a decision tree ensemble, thereby constituting more than one learning model in tandem; See Lin “Introduction” which discloses the paper proposing a Resnet net-based model for pre-screening of OSA diagnosis; See Lin Box 9 & Box 10 which disclose the outputted data including degree of said OSA diagnosis, such as mild, moderate, and severe).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the already-combined disclosure of Lin and Gill with further teachings from Gill regarding the use of multiple machine learning models in tandem or cascaded models, because this allows for multiple nodes or algorithms and adjusting internal derived calculated weights between each of the established node connections by minimizing an error function against actual values during the training process at each node and/or subsequent node (See Gill Par [0100]).
Claim 6 –
Regarding Claim 6, Lin and Gill disclose the method of claim 1 in its entirety. Lin and Gill further disclose a method, wherein:
the generating of the analysis data includes inputting the facial photograph information into a first machine learning model (See Lin Box 1 which discloses that patients may have specific face shapes or structures that affect airway space, leading to an increased risk of OSA, such that specific screening of craniofacial structure, i.e. landmarks, can be considered as an alternative to typical upper airway imaging; See Lin Box 4 which discloses analyzing and comparing some features of craniofacial structure, such that there is a relationship between OSA and some features of craniofacial structure, such as the obvious linear relationship between OSA, neck circumference, mandibular width, and maxillary volume; See Lin Box 8 which discloses in the diagnosis of OSA, focusing on the data of craniofacial structures, i.e. landmark information under BRI, including mandible, temporomandibular, and lateral faces; See Lin Box 5 which discloses identifying, i.e. extracting face key points and then using existing face feature point library to extract and detect them),
generating first analysis data inferring the degree of OSA by the first machine learning model (See Lin Box 1 which discloses the paper proposing a Resnet net-based model for pre-screening of OSA diagnosis; See Lin Box 9 & Box 10 which disclose the outputted data including degree of said OSA diagnosis, such as mild, moderate, and severe),
extracting a plurality of landmark information from the facial photograph information, and generating second analysis data, which is distance information between the landmarks, using the plurality of landmark information (See Lin Box 1 which discloses that patients may have specific face shapes or structures that affect airway space, leading to an increased risk of OSA, such that specific screening of craniofacial structure, i.e. landmarks, can be considered as an alternative to typical upper airway imaging; See Lin Box 4 which discloses analyzing and comparing some features of craniofacial structure, such that there is a relationship between OSA and some features of craniofacial structure, such as the obvious linear relationship between OSA, neck circumference, mandibular width, and maxillary volume; See Lin Box 8 which discloses in the diagnosis of OSA, focusing on the data of craniofacial structures, i.e. landmark information under BRI, including mandible, temporomandibular, and lateral faces; See Lin Box 5 which discloses identifying, i.e. extracting face key points and then using existing face feature point library to extract and detect them; See Lin Box 6 which discloses feature images being extracted for analysis, such that geometric features and texture features, can be used to specify characteristics, such as cervicomental contour area, face/mandibular width (i.e. distance under BRI), and tragion-ramus-stomion angle), and
the inferring of the information about the degree of OSA includes inputting the first analysis data, the second analysis data, and the response data into a fourth machine learning model (It is understood by Examiner that optimizing or determining the amount of learning models to be applied amounts to mere optimization within prior art conditions or through routine experimentation (See MPEP 2144.05(II)) such that specifying the amount of machine learning models beyond two models (i.e. specifying that the model is an ensemble or cascaded machine learning model) is understood to not necessarily hold patentable weight beyond specifying that the model is an ensemble model or comprising more than one learning model in tandem; therefore, see Gill Par [0100] which discloses the potential learning models comprising one or more models, such as in a decision tree ensemble, thereby constituting more than one learning model in tandem), and
inferring the information about the degree of OSA by the fourth machine learning model and generating the inference result (It is understood by Examiner that optimizing or determining the amount of learning models to be applied amounts to mere optimization within prior art conditions or through routine experimentation (See MPEP 2144.05(II)) such that specifying the amount of machine learning models beyond two models (i.e. specifying that the model is an ensemble or cascaded machine learning model) is understood to not necessarily hold patentable weight beyond specifying that the model is an ensemble model or comprising more than one learning model in tandem; therefore, see Gill Par [0100] which discloses the potential learning models comprising one or more models, such as in a decision tree ensemble, thereby constituting more than one learning model in tandem; ).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the already-combined disclosure of Lin and Gill with further teachings from Gill regarding the use of multiple machine learning models in tandem or cascaded models, because this allows for multiple nodes or algorithms and adjusting internal derived calculated weights between each of the established node connections by minimizing an error function against actual values during the training process at each node and/or subsequent node (See Gill Par [0100]).
Claim 7 –
Regarding Claim 7, Lin and Gill disclose the method of claim 1 in its entirety. Lin further discloses a method, wherein:
the machine learning model infers information about a plurality of classes based on a preset range for an apnea-hypopnea index or a respiratory distress index (Without further specifying “classes”, it is understood that any general classification/assigning of the patient to a severity level could constitute a severity “class” under BRI; therefore, see Lin Box 9 & Box 10 which describes the gold standard of OSA including Apnea Hypopnea Index (AHI) ranges and classifying severity of patients’ OSA based on said index).
Claim 8 –
Regarding Claim 8, Lin discloses an apparatus for predicting obstructive sleep apnea, the apparatus comprising:
a memory that stores at least one instruction (See Gill Par [0041]-[0042] which discloses the processes recited throughout Gill being performed via an application, which may be stored on a computer-readable medium, such as memory/data storage, includes programmed instructions for processor to perform said processes/tasks);
a processor that executes the at least one instruction (See Gill Par [0041]-[0042] which discloses the processes recited throughout Gill being performed via an application, which may be stored on a computer-readable medium, such as memory/data storage, includes programmed instructions for processor to perform said processes/tasks), wherein
the processor generates analysis data from facial photograph information of an analysis subject (See Lin Box 2 which discloses the paper proposing a Resnet net-based model for pre-screening of OSA diagnosis based on 2D face recognition of 2D facial photos; See Gill Par [0041]-[0042] which discloses the processes recited throughout Gill being performed via an application, which may be stored on a computer-readable medium, such as memory/data storage, includes programmed instructions for processor to perform said processes/tasks),
stores response data of an OSA screening questionnaire of the analysis subject in the memory (See Gill Par [0041]-[0042] which discloses the processes recited throughout Gill being performed via an application, which may be stored on a computer-readable medium, such as memory/data storage, includes programmed instructions for processor to perform said processes/tasks; See Gill Par [0103] which discloses scan data may be obtained from a storage device, i.e. stored in said storage device in order for retrieval to occur; See Gill Par [0089]-[0091] which discloses collecting subjective data via questionnaire with questions to gather data on symptoms of sleep disorders such that the collected patient input data may be assigned to a patient database, i.e. in storage/memory),
inputs the analysis data and the response data into a pre-trained machine learning model (See Lin Box 2 which discloses the paper proposing a Resnet net-based model for pre-screening of OSA diagnosis; See Lin Box 3 which discloses a machine learning algorithm being used to analyze the human face and Lin Box 11 which states that the analysis is subsequently being fed into a neural network for extraction and classification, such that the analysis data; See Lin Box 8 which discloses the training data used to produce the pre-trained machine learning model that is used for producing the OSA diagnosis) and infers information about the degree of OSA (See Lin Box 1 which discloses the paper proposing a Resnet net-based model for pre-screening of OSA diagnosis; See Lin Box 9 and Lin Box 10 which disclose the outputted data including degree of said OSA diagnosis, such as mild, moderate, and severe), and
transmits the inference result to at least one terminal or outputs the inference result to a display (See Lin Box 10 which discloses displaying the decision result of the model in the form of saliency map, albeit not “transmitted” or a “terminal”, however this limitation has an “or” which means only one of these circumstances has to be met under BRI).
Claims 9-14 –
Regarding dependent claims 9-14, these claims are substantially similar to dependent claims 2-7, but recited for an apparatus performing a method by a processor rather than a method being performed by a processor. Additionally, Lin and Gill effectively disclose the entirety of claim 8 from which dependent claims 9-14 depend and effectively disclose an apparatus comprising a processor for performing the methods/steps recited. Furthermore, independent claim 8 is substantially similar to independent claim 1, but also recited for an apparatus performing a method by a processor rather than a method being performed by a processor. Therefore, claims 9-14 are rejected for the same or substantially similar reasons as claims 2-7, but recited for an apparatus performing a method by a processor, which is additionally met by Lin and Gill, as discussed above. Therefore, claims 9-14 remain rejected under 35 U.S.C. 103 over Lin and Gill for the same or substantially reasons as claims 2-7.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure:
Hogg et al. (U.S. Patent Publication No. 2026/0013787) discloses a method for control system based on a machine learning algorithm, the image of the user and includes determining, by the control system based on the analyzing the image of the user, the sleep score for the user including sleep apnea;
Wilf et al. (U.S. Patent Publication No. 2025/0046453) discloses machine learning image analysis and, more particularly, to face and head-image-based method and system for diagnosing predisposition to a disease such as obstructive sleep apnea;
Zigel et al. (U.S. Patent Publication No. 2015/0351663) discloses a system for determining sleep apnea and apnea-hypopnia index from speech assessment.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to HUNTER J RASNIC whose telephone number is (571)270-5801. The examiner can normally be reached M-F 8am-5:30pm.
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, Shahid Merchant can be reached at (571) 270-1360. 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.
/H.R./Examiner, Art Unit 3684
/KENNETH BARTLEY/Primary Examiner, Art Unit 3684
Prosecution Insights