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 .
Claim Rejections - 35 USC § 112
The following is a quotation of the first paragraph of 35 U.S.C. 112(a):
(a) IN GENERAL.—The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.
The following is a quotation of the first paragraph of pre-AIA 35 U.S.C. 112:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
Claims 18-19 and 26-27 are rejected under 35 U.S.C. 112(a) or 35 U.S.C. 112 (pre-AIA ), first paragraph, as failing to comply with the written description requirement. The claim(s) contains subject matter which was not described in the specification in such a way as to reasonably convey to one skilled in the relevant art that the inventor or a joint inventor, or for applications subject to pre-AIA 35 U.S.C. 112, the inventor(s), at the time the application was filed, had possession of the claimed invention.
Claims 18 and 26 recite inputting data to a generative adversarial network (GAN), convolutional neural network (CNN), long short-term memory neural network model (LSTM), and outputting data based on the inputs. However, the specification is devoid of the governing equations and structure of the models to reasonably convey possession of the claimed invention. As such, the subject matter related to the above models is not adequately described in the specification.
Claims 19 and 27 recite a loss equation for applying categorical cross entropy to classify the lying posture, but lacks detail in the specification. The specification merely recites the claim language, but does not provide or suggest details of what the variables in the equation mean(¶ [0066-71]). As such, the claims lack proper written description and do not reasonably convey possession of the claimed invention.
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 18-34 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 18 and 26 recite a GAN, CNN and LSTM, but is indefinite because the structure or governing equations corresponding to the models are unknown. The models appear to merely receive an input and provide an output, without describing how the input is manipulated to arrive at the output.
Claims 19 and 27 recite “applying categorical cross entropy as a loss function to classify the lying posture of the user”, but is indefinite. It is unclear what the variables for the equation are. As such, one of ordinary skill in the art cannot examine the claim without knowing what the loss function is referring to.
Claims not listed are rejected by virtue of claim dependency.
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.
Claims 18-34 are rejected under 35 U.S.C. 103 as being unpatentable over Ghosh et al. (US 20230145268- Previously cited), hereinafter Ghosh, further in view of Nourani et al. (US 20130090571- Previously cited), hereinafter Nourani, and Alnujaim (Augmentation of Doppler Radar Data Using Generative Adversarial Network for Human Motion Analysis- 2019), hereinafter Alnujaim, and Zhao et al. (Self-Supervised Learning From Multi-Sensor Data for Sleep Recognition- 2020), hereinafter Zhao.
Regarding claim 18, Ghosh teaches an artificial intelligence-based posture discrimination device, method, and non-transitory computer-readable recording medium using body pressure sensors (abstract and ¶[0059,0078], “ present invention may be a system, a method . . . computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present invention.”), the device comprising:
a body pressure sensor array installed on an upper part of a frame of a bed or inside a mattress (¶[0046], “IoT component can manage smart bed 103, which includes the capability of interfacing to various sensors throughout the bed (e.g., position, pressure sensor, etc.).” (emphasis added) “[T]hroughout the bed” viewed with the broadest reasonable interpretation includes sensors that are both in and on the bed), configured to output actual body pressure measurement data of a user touching the mattress;
one or more processors; and
memory storing instructions that, when executed by the one or more processors (¶[0004]), cause the one or more processors to:
discriminate a lying posture of the user based on an output of a neural network ((¶[0022,0051,0058,0062], “generate a position profiling based on time duration,” “patient component 111, through patient component 121, retrieves data associated with a patient,” “AI component 124, generates various patient bed/lying positions based on the digital twin copy of the patient. The bed positions may include the following, non-compliant position, degree of non-compliances, pain modeling in one/many portions of patient, one/more bed bedsores/pressure ulcers, the time period of lying in one position and stress”, “data related to patient A, can include the position of the patient (patient is lying on his back in the middle of the bed,” and “simulates patient movement (on digital twin server 104) based on the medical requirements/parameters.” The excerpts above teach determining the user’s posture and other parameters based on the user’s actual time-series data, generated sample data (digital twin), simulated/predicted patient movement).).
Ghosh fails to teach wherein the body pressure sensor array including a plurality of body pressure sensors arranged in rows and columns and configured to output, over time, actual body-pressure measurement data representing a time series of two-dimensional body-pressure distribution frames of a user touching the frame of the bed or the mattress;
generate sample body-pressure distribution data for the user by applying a Generative Adversarial Network (GAN) to the actual body-pressure measurement data,
wherein the GAN generates the sample body-pressure distribution data by over-sampling the actual body-pressure measurement data;
integrate the sample body-pressure distribution data with the actual body-pressure measurement data to form integrated body-pressure distribution data;
generate a training dataset and a test dataset from the integrated body-pressure distribution data;
train and test a Convolutional-Long Short-Term Memory (CNN-LSTM) neural network using the training dataset and the test dataset, wherein the CNN-LSTM neural network learns and predicts image-type input values in time series; and;
discriminate a lying posture of the user based on an output of the CNN-LSTM neural network.
As recited above, Ghosh fails to teach wherein the body pressure sensor array including a plurality of body pressure sensors arranged in rows and columns and configured to output, over time, actual body-pressure measurement data representing a time series of two-dimensional body-pressure distribution frames of a user touching the frame of the bed or the mattress.
Nourani teaches a method and system of monitoring and preventing pressure ulcers via pressure sensor matrix distributed on or embedded in a bed (abstract and ¶[0032,0071]).
It would have been obvious to one of ordinary skill in the art at the time the invention was effectively filed to have modified the device of Ghosh, such that the body pressure sensor module comprises the plurality of body pressure sensors configured in a three-dimensional structure and a plurality of rows and columns, as taught by Nourani, to aid in monitoring and preventing pressure ulcers.
Ghosh-Nourani fail to teach generate sample body-pressure distribution data for the user by applying a Generative Adversarial Network (GAN) to the actual body-pressure measurement data,
wherein the GAN generates the sample body-pressure distribution data by over-sampling the actual body-pressure measurement data, and
integrate the sample body-pressure distribution data with the actual body-pressure measurement data to form integrated body-pressure distribution data;
generate a training dataset and a test dataset from the integrated body-pressure distribution data.
Alnujaim teaches human motion analysis can be applied to the diagnosis of musculoskeletal diseases, rehabilitation therapies, fall detection, and estimation of energy expenditure (pg. 1, see Objectives). To analyze human motion, deep learning algorithms are most effective approaches (pg. 1, see Objectives). Further, employing a GAN augments the amount of training data, thereby improving the accuracy of human motion recognition (pg. 1, see Objectives). The synthesized data and original data is integrated, to design and train the deep convolutional neural network(DCNN) structure, leading to “improv[ing] the recognition of human motion from 90% to 94%” (pg. 348, “With the combination of original data and synthesized data, the DCNN is designed and trained.”). Alnujaim teaches that “A GAN is a machine learning algorithm designed to produce large amounts of synthesized data that have similar distributions to that of the original data” (pg. 345, ¶[1]).
Therefore, it would have been obvious to one of ordinary skill in the art at the time the invention was effectively filed to have modified the device of Ghosh-Nourani in view of Alnujaim to have generated sample body-pressure distribution data for the user by applying a Generative Adversarial Network (GAN) to the actual body-pressure measurement data, by over-sampling the actual body-pressure measurement data, and integrated the sample body-pressure distribution data with the actual body-pressure measurement data to form integrated body-pressure distribution data; and generate a training dataset and a test dataset from the integrated body-pressure distribution data, to improve the recognition of body pressure measurements when there is insufficient pressure data to discriminate the lying posture.
Ghosh-Nourani-Alnujaim fail to teach train and test a Convolutional-Long Short-Term Memory (CNN-LSTM) neural network using the training dataset and the test dataset, wherein the CNN-LSTM neural network learns and predicts image-type input values in time series; and
discriminate a lying posture of the user based on an output of the CNN-LSTM neural network. It is noted, however, Alnujaim teaches that the actual and synthesized data is used to train and test a DCNN (pg. 348, “With the combination of original data and synthesized data, the DCNN is designed and trained”).
Zhao teaches self-supervised learning from multi-sensor data for sleep recognition comprising pressure sensors to determine postures of the subject (pg. 93918 and figs. 20(c)-(d), “FIGURES 20(c) and 20(d) show the comparison of data dimension reduction in experiment I of the pressure map dataset. Obviously, the correct boundaries of different sleep postures are clearer after using the self-supervised model. In experiment II, due to the small amount of data, we can see a significant aggregation effect. It is difficult to identify the three postures in FIGURE 20(e) when they are fused together, and it is easier to identify the class type of sample points in FIGURE 20(f) when they are close to the same class.”). Posture recognition is based on a CNN and LSTM architecture model (pg. 93908, see Sleep Posture Recognition, and fig. 1, “The third layer uses a standard CNN model to pre-train and expand the data to get Im. The first layer of the downstream task sends the features extracted from the pre-training to BiLSTM”). This combination “extract the temporal features of sleep data without any fitting phenomenon . . . . the input data features are rich and the separability is strong, resulting in better experimental results” (pg. 93915-16, see Result of the Classification of Sleep Posture in Experiment II). Lastly, Zhao contemplates the use of a GAN for image generation and expand on the invention (pg. 93920, see Conclusion).
Therefore, it would have been obvious to one of ordinary skill in the art at the time the invention was effectively filed to have modified the device of Ghosh-Nourani-Alnujaim, to train and test a Convolutional-Long Short-Term Memory (CNN-LSTM) neural network using the training dataset and the test dataset, wherein the CNN-LSTM neural network learns and predicts image-type input values in time series; and discriminate a lying posture of the user based on an output of the CNN-LSTM neural network, as taught by Zhao, to aid in enabling a self-supervised approach to recognizing sleep parameters (see Conclusion on pg. 93920).
Regarding claims 19 and 27, Zhao teaches wherein training the CNN-LSTM neural network comprises applying categorical crossentropy as a loss function to classify the lying posture of the user (pg. 93911, “This section introduces the loss function of the pre-training process, which is determined by KL divergence and cross entropy function”).
Regarding claims 20 and 28, Ghosh teaches wherein the lying posture discriminated by the one or more processors comprises at least a supine posture (¶[0059], “can include the position of the patient (patient is lying on his back in the middle of the bed),”(emphasis added) ).
Regarding claims 21 and 29, Ghosh-Nourani-Alnujaim-Zhao teach wherein each of the two-dimensional body-pressure distribution frames comprises pressure values arranged according to respective positions of the plurality of body pressure sensors in the rows and columns (¶[0022,0046,0051,0058,0062] of Ghosh, “generate a position profiling based on time duration,” “patient component 111, through patient component 121, retrieves data associated with a patient,” “AI component 124, generates various patient bed/lying positions based on the digital twin copy of the patient. The bed positions may include the following, non-compliant position, degree of non-compliances, pain modeling in one/many portions of patient, one/more bed bedsores/pressure ulcers, the time period of lying in one position and stress”, “data related to patient_A, can include the position of the patient (patient is lying on his back in the middle of the bed,” and “smart bed 103, which includes the capability of interfacing to various sensors throughout the bed (e.g., position, pressure sensor” (emphasis added)).
Regarding claims 22 and 30, Ghosh teaches further comprising a storage device comprising a database associated with the user, wherein the instructions, when executed by the one or more processors, further cause the one or more processors to store and manage, in the database, user information data comprising at least the corresponding user’s lying posture information data discriminated by the posture discrimination module (¶[0041], “storage device capable of storing data and configuration files that can be accessed and utilized by server” and “Database 116 may store information associated with, but is not limited to, knowledge corpus, patient medical history, IoT devices, smart bed profile and setting, modeling of patient, patient activities and home automation routines”).
Regarding claims 23 and 31, Ghosh teaches further comprising a communication interface configured for wired or wireless communication with an external terminal or a server, wherein the instructions, when executed by the one or more processors, further cause the one or more processors to transmit, via the communication interface to the external terminal or the server, user information data comprising at least the corresponding user’s lying posture information data discriminated by the posture discrimination module (fig. 2 and ¶[0038], “Server 110 and/or digital twin server 104 can be a standalone computing device, a management server, a web server, a mobile computing device, or any other electronic device or computing system capable of receiving, sending, and processing data”).
Regarding claims 24 and 32, Ghosh teaches wherein the external terminal or the server (110 of Ghosh) allows the user information data to be displayed on a display screen thereby enabling a manager to check the user information data of the user visually (¶[0079] of Ghosh, “Display 409 provides a mechanism to display data to a user”) the user information data being transmitted from the communication module through a pre-installed specific application service (¶[0041], application service of Ghosh), but fails to teach displaying the user information data of at least the corresponding user’s actual body pressure distribution data, or the corresponding user’s lying posture information data.
Nourani teaches that the display can output the body pressure data with the corresponding lying posture information data (fig. 10 and [0081]).
As such, it would have been obvious to one of ordinary skill in the art at the time the invention was effectively filed to have modified the device of Ghosh-Nourani-Alnujaim-Zhao, such that the user information data of at least the corresponding user’s actual body pressure distribution data, or the corresponding user’s lying posture information data, is displayed, as taught by Nourani, because Ghosh requires display user information, but fails to provide details, and Nourani teaches that user information can be the corresponding user’s actual body pressure distribution data, or the corresponding user’s lying posture information data.
Regarding claims 25 and 32, Ghosh teaches wherein the external terminal or the server comprises a database associated with the user and is configured to store and manage, in the database, the user information data transmitted via the communication interface through a pre-installed application service (¶[0041], “database 116 resides on server 110 . . . database 116 may reside elsewhere within patient mobility environment 100, provided that patient component 111 has access to database 116” and “invention may be a system, a method, and/or a computer program product at any possible technical detail level of integration” (emphasis added) indicating that computer program products (pre-installed specific application service) can be integrated in the various devices, computers, etc., as described including the computer(s) that communicates with the server/database. It is noted, the specification of the present invention does not specify what the specific application service is.).
Response to Arguments
Applicant's arguments filed 07/05/2026 have been fully considered but they are not fully persuasive.
Applicant’s arguments with respect to claims 1-17 have been considered but are moot because they have been replaced by claims 18-34 that require new grounds of rejection.
It is noted that claims related to the previous 35 U.S.C. 112(a) were amended but not addressed. Since the invention, as a whole, relies on those teachings, response from Applicant is required.
Applicant’s arguments related to Ghosh lacking teachings related to GAN and CNN-LSTM use, on page 10 of the Remarks, is correct. However, Applicant fails to argue the combination of references because Ghosh is not relied upon for those teachings.
Applicant’s arguments related to Nourani are moot because of amendments to the claims and newly applied art.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Gangwar teaches ML model can include a long term short term memory (LSTM) based model having culmination of logistic regression model and neural network based bi-directional LSTM cells, wherein the knowledgebase is used to train LSTM neural net using categorical cross entropy as loss function and an optimizer, wherein the ML model facilitates supervised learning. US 20220207066
Albero teaches one or more video cameras may be configured to record videos at the retail banking physical location and send the videos to the electronic monitoring platform. An AI engine (e.g., using a convolutional neural network, generative adversarial network, or any other type of machine learning algorithm) associated with the electronic monitoring platform may analyze recordings from the video cameras and determine movement patterns (e.g., gaits, paces, and postures) associated with the individuals at the physical location. US 20220415146
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MARTIN NATHAN ORTEGA whose telephone number is (571)270-7801. The examiner can normally be reached M-F 7:10 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, Robert (Tse) Chen can be reached at (571) 272-3672. 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.
/MARTIN NATHAN ORTEGA/ Examiner, Art Unit 3791 /TSE CHEN/Supervisory Patent Examiner, Art Unit 3791
Prosecution Insights