DETAILED ACTION
Acknowledgement
This final office action is in response to the amendment filed on 10/14/2025.
Status of Claims
Claims 1-4, 10-13, 17, and 22-24 have been amended.
Claims 1-15, 17, and 21-24 are now pending.
Response to Arguments
Applicant's arguments filed on 10/14/2025 regarding the 35 U.S.C. 101 and 103 rejections of claims 1-15, 17, and 21-24 have been fully considered. The Applicant argues the following.
(1) As per the 101 rejection, the Applicant argues, in summary, that (i) claim 1 is not directed toward Certain Methods of Organizing Human Activity nor Mental Processes because the steps cannot practically be performed in a human mind (ii) claims recite an improved technique for training a machine learning model for accident prediction and include several features that integrate any alleged judicial exception into a practical application, such as improving the functioning of a computer; and (iii) the claims recite an unconventional combination of operations and data structures that provides non-routing results in the field of process controls and thus the claims provide an inventive concept.
The Examiner respectfully disagrees. The Examiner maintains the position that claim 1 as amended recite abstract elements as listed in Step 2A(1) that are directed to the abstract grouping of Mental Processes. The abstract elements describe a process of collecting, validating, and analyzing assessment data to determine an environment, health, or task violation, generating recommendations and alerts, and display results, which can be practically performed in the human mind with pen and paper through physical observation, evaluation, and judgment. A human person can visually observe a workplace environment, analyze data via mental math, and make judgments on threshold violations, recommendations, and generate alerts. Mental Processes include claims directed to collecting information, analyzing it, and displaying certain results of the collection and analysis even if they are claimed as being performed on a computer. Per MPEP 2106.04(a), a claim recites a judicial exception when the judicial exception is “set forth” or “described” in the claim.
The Examiner also submits that the additional elements recited in the claims and listed in Steps 2A(2) and 2B do not integrate the abstract idea into a practical application nor provide an inventive because the additional elements do not reflect an improvement in the functioning of a computer or an improvement in another technology or technical field. The claims do not reflect a new and improved way of training or retraining a machine learning model based on the limitations recited in the claims and the support of the Applicant’s specification. The additional elements recited in the claim reflect mere use of computer-based technology that improves upon an abstract process of analyzing data and communicating results.
The Examiner does not find that the additional elements reflect an unconventional combination of operations and data structures that provides non-routine results in the field of process controls. The claims provide results of generating and displaying alerts/notifications informing a user of a violation and do not provide a technical improvement in the field. Therefore, the 35 U.S.C. 101 rejection is maintained.
(2) As per the 103 rejections, the Applicant argues, in summary, that Barak fails to teach or suggest the recited features of amended independent claim 1 of "training...the machine learning model on the wearable, the near accident, the at least one violation, and one or more acceptable ranges associated with the collected task data or the real-time health vitals, wherein the one or more acceptable ranges are defined uniquely for the at least one task " and deSa and Saripalli fail to cure these deficiencies.
The Examiner respectfully disagrees. The Examiner submits that while Barak alone does not teach all of the limitations of amended claim 1, Barak in view of deSa teach the limitations as shown in the updated claim mapping below. Therefore, the 35 U.S.C. 103 rejection is maintained.
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 § 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 1-15, 17, and 21-24 are rejected under 35 U.S.C. 101 because the claimed invention, “Systems and Methods For Next Generation Connected-Worker Solutions For Occupational Safety, Health, and Productivity”, is directed to an abstract idea, specifically Mental Processes and Certain Methods of Organizing Human Activity, without significantly more. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the additional elements individually or in combination provide mere instructions to implement the abstract idea on a computer.
Step 1: Claims 1-15, 17, and 21-24 are directed to a statutory category, namely a process (claims 1-9 and 21-24), a machine (claims 10-15), and a manufacture (claim 17).
Step 2A (1): Independent claims 1, 10, and 17 are directed to an abstract idea of Mental Processes, based on the following claim limitations: “…identifying one or more alerts…; generating,…, by inputting assessment data…, a recommendation for a wearable for at least one user…to wear while performing at least one task, wherein (i) the assessment data is provided… (ii) the assessment data includes environment information corresponding to at least one environment of the one or more devices, task information corresponding to at least one task associated with the one or more devices, and user information corresponding to at least one user; validating,…, the environment information and the user information of the assessment data based on additional data received …; capturing,…, collected task data… of the at least one user while the at least one user performs the at least one task; generating,…, one or more alerts indicating that the at least one user is associated with at least one violation by comparing the collected task data to one or more thresholds, wherein the at least one violation comprises a health violation and the one or more thresholds comprise a health threshold, and wherein comparing the collected task data to one or more thresholds comprises comparing the…health vitals of the at least one user to the health threshold to determine the health violation; and displaying,…, the one or more alerts including information indicating the at least one violation; detecting,…, a near accident associated with the at least one violation.”. These claims describe a process of collecting, validating, and analyzing assessment data to determine an environment, health, or task violation, generating recommendations and alerts, and display results, which can be practically performed in the human mind with pen and paper through physical observation, evaluation, and judgment. Dependent claims 2, 3, 6, 8, 9, 11, 12, 15, and 21-22, and 24 further describe the analysis of assessment data to determine violations, provide recommendations, and display results. These limitations, under the broadest reasonable interpretation, fall within the abstract groupings of Mental Processes which include concepts performed in the human mind such as observations, evaluations, judgments, and opinions. Mental Processes include claims directed to collecting information, analyzing it, and displaying certain results of the collection and analysis even if they are claimed as being performed on a computer. Dependent claims 4, 5, 13, 14, and 21 limitations of “determining,…, whether the at least one task has been completed; and in response to determining that the at least one task has been completed, displaying,…, at least one record corresponding to the at least one task…(claims 4 and13); assigning,…, at least one task to the at least one user, the assigning based on the environment information, the task information, and the user information (claims 5 and 14); determining that the at least one task was not completed; and automatically updating a…record to indicate equipment to be used to perform a second task, wherein the second task comprises a repeat of the at least one task, wherein the equipment is selected in order to perform the second task without causing a second health violation while performing the second task (claim 21)” reflect Certain Methods of Organizing Human Activity as the monitoring of task completion or non-completion of a user and assigning tasks and equipment to users to perform tasks are actions that manages a person’s behavior and reflects instructions for a person to follow. Certain Methods of Organizing Human Activity encompasses managing personal behavior or relationships or interactions between people including social activities, teaching, and following rules or instructions. Certain Methods of Organizing Human Activity can encompass the activity of a single person (e.g. a person following a set of instructions), activity that involve multiple people (e.g. a commercial interaction), and certain activity between a person and a computer (e.g. a method of anonymous loan shopping). Therefore, claims 1-15, 17, and 21-24 are directed to an abstract idea and are not patent eligible.
Step 2A (2): This judicial exception is not integrated into a practical application. In particular, claims 1-7, 10-15, 17, and 21-24 recite additional elements of “a computer, …by one or more processors, one or more devices, machine learning model, one or more additional devices, capturing, using the wearable, real-time health vitals…, user interfaces of one of one of the devices; training, by the one or more processors, the machine learning model; equipment; wherein the one or more devices include one or more sensors, one or more mobile devices, one or more internal computing systems, and/or one or more external computing systems; a computer system…comprising a memory having processor-readable instructions stored therein; one or more processors configured to access the memory and execute the processor-readable instructions, which when executed by the one or more processors configures the one or more processors to perform a plurality of functions; a database, transmitting…to an external computing system, wherein the external computing system is configured; transmitting…to an internal computing system, wherein the internal computing system is configured; wherein the machine-learning model comprises a transformer machine-learning network and wherein the computer- implemented method further comprises: training the transformer machine-learning network using historical health vitals and historical health acceptable ranges; and updating the transformer machine-learning network based on the real-time health vitals; and causing one or more graphical representations of the predictive preview feature to be rendered on a graphical user interface of a mobile device of the at least one user, wherein the mobile device is one of the one or more devices. ”. These additional elements do not integrate the abstract idea into a practical application because the claims do not recite (a) an improvement to another technology or technical field and (b) an improvement to the functioning of the computer itself and (c) implementing the abstract idea with or by use of a particular machine, (d) effecting a particular transformation or reduction of an article, or (e) applying the judicial exception in some other meaningful way beyond generally linking the use of an abstract idea to a particular technological environment. These additional elements evaluated individually and in combination are viewed as computing, data gathering, and display devices that are used to perform the abstract idea. Limitations that recite mere data gathering and mere instructions to implement an abstract idea on a computer or merely uses a computer as a tool to perform an abstract idea are not indicative of integration into a practical application (see MPEP 2106.05(f)). Also, limitations that amount to merely indicating a field of use or technological environment (e.g. machine learning) in which to apply a judicial exception do not amount to significantly more than the exception itself, and cannot integrate a judicial exception into a practical application (see MPEP 2106.05(h)). Therefore, claims 1-15, 17, and 21-24 do not include individual or a combination of additional elements that integrate the judicial exception into a practical application and thus are not patent eligible.
Step 2B: The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception. Claims 1-7, 10-15, 17, and 21-24 recite additional elements of “a computer, …by one or more processors, one or more devices, machine learning model, one or more additional devices, capturing, using the wearable, real-time health vitals…, user interfaces of one of one of the devices; training, by the one or more processors, the machine learning model; equipment; wherein the one or more devices include one or more sensors, one or more mobile devices, one or more internal computing systems, and/or one or more external computing systems; a computer system…comprising a memory having processor-readable instructions stored therein; one or more processors configured to access the memory and execute the processor-readable instructions, which when executed by the one or more processors configures the one or more processors to perform a plurality of functions; a database, transmitting…to an external computing system, wherein the external computing system is configured; transmitting…to an internal computing system, wherein the internal computing system is configured; wherein the machine-learning model comprises a transformer machine-learning network and wherein the computer- implemented method further comprises: training the transformer machine-learning network using historical health vitals and historical health acceptable ranges; and updating the transformer machine-learning network based on the real-time health vitals; and causing one or more graphical representations of the predictive preview feature to be rendered on a graphical user interface of a mobile device of the at least one user, wherein the mobile device is one of the one or more devices. ”. These additional elements evaluated individually and in combination are viewed as mere data gathering and instructions to apply or implement the abstract idea on a computer and merely indicates a field of use or technological environment in which to apply a judicial exception. The use of machine learning and trained models/algorithms are considered instructions to apply or implement a model on a computer. Applying an abstract idea on a computer and/or generally linking the use of the abstract idea to a particular technological environment does not integrate a judicial exception into a practical application or provide an inventive concept (see MPEP 2106.05 (f) and (h)). Therefore, claims 1-15, 17, and 21-24 do not include individual or a combination of additional elements that are sufficient to amount to significantly more than the judicial exception and thus are not patent eligible.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
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 1-15, 17, 21-22, and 24 are rejected under 35 U.S.C. 103 as being unpatentable over Barak et al. (US 2020/0202471 A1) in view of deSa et al. (US 2021/0057093 A1).
As per claims 1, 10, and 17 (Currently Amended), Barak teaches a computer-implemented method comprising (Barack e.g. Methods and systems for providing different real-time safety information at a plurality of locations within an industrial environment [0007]. Systems, methods, and devices that initiate safety related actions based on determined risk of a task [0002]. The suggested systems and methods continuously identify hazards by choosing relevant data originating from different sources, calculate the current risk score, and initiate actions to prevent personal accidents and process accidents [0006].); A computer system for identifying one or more alerts relating to a connected system, the computer system comprising: a memory having processor-readable instructions stored therein; and one or more processors configured to access the memory and execute the processor-readable instructions, which when executed by the one or more processors configures the one or more processors to perform a plurality of functions, including functions for (Barak e.g. System 100 may represent a computer-based system that includes computer system components,...memory devices, and/or internal network(s) connecting the components [0037]. System 100 may include at least one sensing device 105 that may (or may not) be associated with employee 110, a server 115 operatively connected to a database 120, and an output unit 125 associated with the industrial environment [0038]. FIG. 2 is a block diagram of example configurations of server 115 and sensing device 105 [0047]. Processing device 202, shown in FIG. 2, may include at least one processor configured to execute computer programs, applications, methods, processes, or other software to perform embodiments described in the present disclosure [0048].); A non-transitory computer-readable medium containing instructions for identifying one or more alerts relating to a connected system, the instructions comprising (Barak e.g. Non-transitory computer-readable storage media may store program instructions, which are executed by at least one processing device and perform any of the methods described herein [0008].):
Barak teaches generating, by one or more processors and by inputting assessment data to a machine learning model, a recommendation for a wearable for at least one user of one or more devices to wear while performing at least one task (Barak e.g. In one example, server 115 may be a cloud server that processes data received from one or more sensing devices (e.g., sensing devices 105A-105E) and processes the data to determine a risk score of a task and/or to identify hazards in the industrial environment [0040]. Server 115 may also process the received data to determine recommendations for preventing accidents [0040]. System 100 may use various machine learning or deep learning techniques to determine the value of the risk score using the safety related information, a plurality of rules, and a plurality of factors [0067]. When the task risk score of the scheduled task is above the first threshold, the processing device may issue a notice prohibiting the execution of the scheduled task (block 508) and provide recommendations to minimize the risk of the scheduled task (block 510) [0084]. Process 500 continues when the processing device provides recommendations for a scheduled task (block 516). In one embodiment, the recommendations for a scheduled task may include checklists, relevant warnings, suggested tools, and more [0085]. FIG. 12 illustrates method 1200 for providing real-time safety information at a plurality of locations within an industrial environment [0132]. Step 1250 includes processing the common events overview report to identify a safety-related threat affecting at least one of the plurality of tasks. The safety-related threat may optionally be a threat which is identifiable based (at least partly) on real-time data, which is collected during the execution of a task, and not available previously. For example, the data collected at the previous steps may indicate that a task which takes place at a lab will continue until 17:00 instead of 16:30. Since another task which is scheduled for 16:45 at the labs involves laser emission, there is an eye safety hazard for the employees performing the earlier task. Such a hazard may be resolved in several ways (e.g., postponing the second task, instructing employees of the earlier task to wear laser safety goggles, etc.) in the following steps [0137].), wherein (i) the assessment data is provided by one or more devices, and (ii) the assessment data includes environment information corresponding to at least one environment of the one or more devices, task information corresponding to at least one task associated with the one or more devices, and user information corresponding to the at least one user; (Barack e.g. Embodiments of the present disclosure further include obtaining safety-related information associated with the task scheduled to take place in the industrial environment. The safety-related information may include work procedures associated with the task (e.g., the required safety measures for the task, the minimum number of people required to complete the task, etc.), information associated with an employee assigned to the scheduled task (e.g., information about an employee's current shift and previous shifts, information about the employee's qualifications and seniority, relevant employee's health information such as allergies, etc.), information associated with a location of the scheduled task (e.g., details of other tasks scheduled to take place at a same area, safety restrictions applied to the location, etc.),...information associated with the industrial environment (e.g., infrastructure blueprints, machinery inventory, material inventory, general regulations and specific procedures, a risk analysis plan, etc.), and more [0033].)
Barak teaches validating, by the one or more processors and using the machine learning model, the environment information and the user information of the assessment data based on additional data received from one or more additional devices other than the one or more devices; (Barack e.g. The present disclosure further includes determining first synergy data safety-related information and task characteristics and determining second synergy data from the safety-related information and the real-time information [0035]. As used herein, the term "determining synergy data" refers to a process of cross-reference information from multiple sources and identifying events that may be unidentifiable when considering information from each source separately [0035]. Process confirmation module 308 may use information collected during task execution and confirm that the design integrity, the operational integrity, and the technology integrity comply with work process procedures [0060]. Optionally, the visual representation of the actual execution of the threatened task may be generated based on the task execution information from the plurality of sensors in the industrial environment and the debriefing-responses of the first employee to the debriefing questionnaire [0143]. The debriefing is used to collect information which is not easily available in other means, relating to what went as planned in the task, which deviations from the original plans were, what were the reasons and implications of such deviations, what risk factors did the employee identify during the execution of the task, and so on [0167]. System 100 may analyze data acquired by a plurality of sensing devices 105 to determine a risk score of a task and/or to identify hazards in the industrial environment [0039]. System 100 may use various machine learning or deep learning techniques to determine the value of the risk score using the safety related information, a plurality of rules, and a plurality of factors. The plurality of factors may include industry task risk analysis factors, environmental factors, timing factors, environmental risk factors, task statistics factors, and more [0067]. To determine the type of the hazard, system 100 may use artificial intelligence (AI) and machine learning algorithms [0075].)
Barak teaches capturing, by the one or more processors and using the wearable, collected task data and real-time health vitals of the at least one user while the at least one user is performing the at least one task; (Barack e.g. The methods and systems includes… receiving real-time sensor information from a plurality of sensors in the industrial environment, wherein the real-time sensor information is obtained from at least three different types of sensors selected from a group consisting of: (a) a plurality of cameras located in the industrial environment, (b) a plurality of communication devices of employees in the industrial environment, (c) wearable sensors of employees in the industrial environment, (d) operational technology (OT) sensors, (e) environmental sensors, and (f) sensors associated with working tools; receiving task execution modification information for at least some of the plurality of tasks, including at least three of: (a) detected changes in performances of an employee assigned to the task, (b) detected changes in planned locations of the task, (c) detected changes in tools expected to be used in the task, (d) detected changes in materials expected to be used in the task, (e) detected changes in an expected start time of the task, (f) detected changes in expected duration of the task, and (g) detected changes in an expected weather during the task;…[0007]. Sensing device 105 may include a wearable device, such as a smart helmet 105B, smart protective gear, smart glasses, a clip-on camera, etc. [0039]. In another example, smart helmet 105B may include a heart-rate sensor 228 for capturing an employee heart rate [0053]. Parameters sensed by such sensors integrated into the first computer may include parameters relating to the employee itself (e.g., location, body temperature), to the environment of the employee (e.g., ambient temperature, atmospheric contents, light level), to the execution of the task (e.g., if the task involves using of the first computer, if the first computer can connect to machines or tools used or affected in the first task) [0170].)
Barak teaches generating, by the one or more processors, one or more alerts indicating that the at least one user is associated with at least one violation by comparing the collected task data to one or more thresholds (Barak e.g. Embodiments of the present disclosure further include determining a predicted risk score of the scheduled task and determining a change in the risk score of the task. The term "risk score" refers to a score that can be assigned based on comparing synergy data to a risk predictor model. A risk score can have a standard value (e.g., a number) or a multi-value threshold (e.g., a line on a graph) [0036]. If a risk score is greater than a reference risk score, there is increased likelihood that an undesirable event that may involve damage (e.g., physical damages) to workers or machines will occur during or after the task. The system may receive real-time information and update the risk score based on events detected using the real-time information [0036]. FIG. 1 shows an example of a system 100 for analyzing information collected from an industrial environment. System 100 enables obtaining safety related information associated with a task scheduled to take place in the industrial environment. In another embodiment, system 100 enables obtaining real-time information indicative of human error of at least one employee associated with the task [0037]. System 100 may analyze data acquired by a plurality of sensing devices 105 to determine a risk score of a task and/or to identify hazards in the industrial environment [0039]. Real-time intervention-system 100 may detect that a change to the operating environment or a control indicator is not as expected ( e.g., machine operational status, weather, other tasks, worker specific risk profile) or a combination of factors creates a risk score that is unacceptable [0059]. System 100 may initiate a remedial action, such as triggering real-time alerts, preventing the task from being performed by shutting down connected machines, or making the task paused or locked [0059].),
Barak teaches displaying, by the one or more processors, the one or more alerts on one or more user interfaces of one of the one or more devices, the one or more alerts including information indicating the at least one violation; (Barak e.g. When the actual risk score of a task is above a certain threshold, the system may initiate a remedial action to prevent a work accident [0036]. Examples of remedial actions, include transmitting location-based warning messages to employees, displaying the detected hazards on a personalized map, performing an automatic shutdown, and creating customized inspection tour based on the detected locations of the plurality of hazards [0036]. Output unit 125 may display identified real-time hazards and potential hazards on a personalized map together with visual indicators of the hazard's severity and the hazard's type. Output unit 125 may be part of an employee station [0042]. System 100 may initiate a remedial action, such as triggering real-time alerts, preventing the task from being performed by shutting down connected machines, or making the task paused or locked [0059].)
Barak teaches detecting, by the one or more processors, a near accident associated with the at least one violation; and (Barak e.g. The suggested systems and methods continuously identify hazards by choosing relevant data originating from different sources, calculate the current risk score, and initiate actions to prevent personal accidents and process accidents [0006]. The method and systems include…receiving task execution modification information for at least some of the plurality of tasks, including at least three of: (a) detected changes in performances of an employee assigned to the task, (b) detected changes in planned locations of the task, (c) detected changes in tools expected to be used in the task, (d) detected changes in materials expected to be used in the task, (e) detected changes in an expected start time of the task, (f) detected changes in expected duration of the task, and (g) detected changes in an expected weather during the task; [0007]. In one embodiment, the characteristic of the task may include at least one of the following: estimated start time of the task, identity of employees expected to participate in the task, expected duration of the task, potential accidents associated with the task, potential accidents associated with the identity of employees, types of materials expected to be used in the task, and types of tools expected to be used in the task [0032]. Embodiments of the present disclosure further include obtaining real-time information indicative of human error of at least one employee associated with the task. In order to know that at least one employee made an error, the system may compare the obtained real-time information with the work procedures and/or with a predetermined behavior baseline for each employee associated with the task to determine if a deviation exists [0034]. System 100 enables obtaining safety related information associated with a task scheduled to take place in the industrial environment. In another embodiment, system 100 enables obtaining real-time information indicative of human error of at least one employee associated with the task [0037]. system 100 may analyze data acquired by a plurality of sensing devices 105 to determine a risk score of a task and/or to identify hazards in the industrial environment [0039]. To determine the type of the hazard, system 100 may use artificial intelligence (AI) and machine learning algorithms. The types of the hazards may include electrical hazards (e.g., frayed cords, missing ground pins, and improper wiring); machinery-related hazards (e.g., exposed moving machinery parts, and safety guards removed); tripping hazards (e.g., cords running across the floor, and wet floor); height-related hazards (e.g., unsafe ladders, scaffolds, roofs, and any raised work area); biological hazards (e.g., fungi/mold, insect bites, animal and bird droppings); physical hazards (e.g., exposure to radiation, extreme temperatures, and noise); chemical hazards (e.g., spilled liquids, exposure to toxic fumes, explosive chemicals not stored properly, and more) [0075].)
Barak teaches…using the machine learning model based on wearable, the near accident, and the at least one violation (Barak e.g. System 100 may use various machine learning or deep learning techniques to determine the value of the risk score using the safety related information, a plurality of rules, and a plurality of factors. Rules include specific machine learning derived rules, location rules, worker risk analysis rules, policy rules, best practice rules, regulation rules, and more [0067]. Safety related information include wearable sensor data and past safety incidents included in the historical safety related information [0068]. In some embodiments, pre-task planning module 302 may determine the value of the risk score using past data, industry statistics, and operational parameters to predict the likely range of parameters that are likely to be present. For example, pre-task planning module 302 may predict the systematic and specific risk for each task at the planned time, place, worker, and activity scenarios [0067]. If a risk score is greater than a reference risk score, there is increased likelihood that an undesirable event that may involve damage (e.g., physical damages) to workers or machines will occur during or after the task. The system may receive real-time information and update the risk score based on events detected using the real-time information [0036].)
While Barak teaches collecting real-time health vitals, establishing task risk thresholds, and identifying violations as shown above, Barak does not explicitly teach wherein the at least one violation comprises a health violation and the one or more thresholds comprise a health threshold, and wherein comparing the collected task data to one or more thresholds comprises comparing the real-time health vitals of the at least one user to the health threshold to determine the health violation.
However, deSa teaches wherein the at least one violation comprises a health violation and the one or more thresholds comprise a health threshold, and wherein comparing the collected task data to one or more thresholds comprises comparing the…health vitals of the at least one user to the health threshold to determine the health violation; (deSa e.g. The present application relates generally to a monitoring system to assist with aging-in-place for elderly individuals and patients with chronic diseases either living at home, senior living or assisted living facilities. The system and methods may be used to identify and track common daily human activities, instrumental daily living activities, the patient or elderly personal physical status in the home or at senior and assisted living facilities. Anomalies to patterns can be determined by identifying disruptions in previously established patterns (Abstract). A method comprises: receiving at least one measurement from at least one sensor; determining an activity of a patient based on the received at least one measurement; and generating alerts when trends deviate from a set of pre-defined thresholds [0007]. Sensor 3a collects data related to respiratory rate (RR) and heart rate (HR) measurements using a set of algorithms. Sensor 3a measures the mechanical aspects of respiratory activity through chest wall movement, and of heart activity through movements produced by ejection of blood from the heart into the great vessels. Accordingly, the system can produce a set of vital signs on an interval basis or continuously (Fig. 1 and [0023]). The individual can set thresholds based on personal behavioral experiences or the system can set a default range which improves with time [0037]. The example of FIG. 3 shows an embedded, layered system of information and visualization. The dashboard's layered system provides insight into daily routine of living activities, history of specific activities and related anomalies, and vital signs of heart rate (HR) and respiratory rate (RR) as well as vital sign histories [0037]. Also provided is a personal thresholds alarm configuration tab or icon 340 via which the user can personalize alarm thresholds for activities or vitals [0044]. An algorithm creates an optional layered alarm system of user-tailored alerts via the dashboard from low-threshold signals to caregivers and high-threshold alerts to healthcare professionals [0037].)
The Examiner submits that before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify Barak’s real-time safety information system to include determining health violations from health vitals and thresholds as taught by deSa in order to unobtrusively monitor the functional status and safety of patients [and/or workers] over time for earlier identification of exacerbations or decline beyond clinically important thresholds (deSa e.g. [0045]-[0046]).
While Barak teaches using a machine learning model based on the wearable, near accident, and at least one violation data, Barak does not explicitly teach training the machine learning model based on one or more acceptable ranges associated with the collected task data or the real-time health vitals, wherein the one or more acceptable ranges are defined uniquely for the at least one task.
However, deSa teaches training, by the one or more processors, the machine learning model based on one or more acceptable ranges associated with the collected task data or the real-time health vitals, wherein the one or more acceptable ranges are defined uniquely for the at least one task; (deSa e.g. A system for determining an activity of a patient comprises at least one processor and at least one memory including computer program code, the at least one memory and the computer program code configured to, with the at least one processor, cause the system to: receive at least one measurement from at least one sensor; and determine the activity of the patient based on at least one of: (i) rule-based heuristics, (ii) training data from a home of one or more patients, and (iii) the received at least one measurement from the at least one sensor [0006]. According to another aspect, a method comprises: receiving at least one measurement from at least one sensor; determining an activity of a patient based on the received at least one measurement; and generating alerts when trends deviate from a set of pre-defined thresholds [0007]. FIG. 1 is a block diagram of an exemplary integrated and interoperable contactless (meaning not requiring direct contact with the monitored person) sensor technology system with embedded microprocessors and a gateway that communicates with a server where processing develops an informational and actionable dashboard [0011]. Several standard machine learning methods are applied and non-standard models to develop time sequences 212, feature generation 215, activity classification 220, activity discovery 225, and human subject attribution 230 before the output 235 is developed that is related to routine activities 240, 250 and anomaly detections 232 [0036]. An algorithm creates an optional layered alarm system of user-tailored alerts via the dashboard from low threshold signals to caregivers and high-threshold alerts to healthcare professionals [0037]. To train classification algorithms, time-stamped ground truth data is collected at the time of installation based on activation of alternating current electrical devices and water fixtures, plus scripted human activities (which may involve the monitored individual or others). In some embodiments, a monitored individual may wear an accelerometer, gyroscope, and/or radio frequency identification tag to compile additional ground truth model training data (for a period of two weeks or less) [0041]. In some embodiments, data from multiple monitored individuals in different homes may be compiled to further train/tune classification models for improved classification accuracy. This training period typically lasts fewer than 14 days. In some embodiments, additional ground truth data is gathered through periodic interaction with monitored individuals [0041]. These algorithms are distinct due to the nature of the data source that predicts a highly validated behavioral activities of daily living of a human subject along with RR/HR. It is unique to obtain one million observations on human subject's real core activities of daily living in order to train the machine to predict daily human activities [0041]. Neural network (specifically Long Short-Term Networks) and deep learning models are continually updated with time series inputs of labeled activities, mobility measures, discovered activities sequences, and vital signs. These models identify significant changes in these inputs over time, allowing the recognition of anomalous activity on the part of the monitored individual ([0040] and [0053]).)
The Examiner submits that before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify Barak’s real-time safety information system’s machine learning process to include training and updating the machine learning model as taught by deSa in order to unobtrusively monitor the functional status and safety of patients [and workers] over time for earlier identification of exacerbations or decline beyond clinically important thresholds (deSa e.g. [0045]-[0046]).
As per claims 2 and 11 (Currently Amended), Barak in view of deSa teach the computer-implemented method of claim 1 and the computer system of claim 10, Barak teaches further comprising analyzing the collected task data to determine whether the collected task data is within an acceptable range (Barak e.g. FIG. 12 illustrates method 1200 for providing real-time safety information at a plurality of locations within an industrial environment [0132]. Consistent with the systems and methods described above, it is noted that data collected from employees (as well as other data collected throughout method 1200) may be used for generating visual representation indicative of differences between the planning of tasks to what actually happened during execution [0143]. Information pertaining to the actual execution may be represented as a "blue line" or similar representation, while information pertaining to the expected execution may be represented as a "black line" or similar representation [0143]. System 100 may use machine learning algorithms to analyze the data and to provide recommendations to minimize the gaps between the "blue line" and the "black line", or to assist workers, management and/or safety supervisors to process the differences between the blue line and the black line (also referred to as the "performance gap") [0160].)
As per claims 3 and 12 (Currently Amended), Barak in view of deSa teach the computer-implemented method of claim 2 and the computer system of claim 11, Barak teaches the method further comprising: in response to determining that the collected task data is not within the acceptable range, displaying, by the one or more processors, a notification on the one or more user interfaces of the one or more user interfaces of one of the one or more devices, the notification indicating that the collected task data is not within the acceptable range (Barak e.g. A user interface (UI) tool is provided, to identify the gaps between the "blue line" and the "black line", according to the HOP principle. The UI tool may also identify possible errors, dangerous latent conditions, and weak defenses [0160]. System 100 may use machine learning algorithms to analyze the data and to provide recommendations to minimize the gaps between the "blue line" and the "black line", or to assist workers, management and/or safety supervisors to process the differences between the blue line and the black line (also referred to as the "performance gap") [0160]. FIG. 13 illustrates an example user interface showing the "blue line" and the "black line" according to the HOP principle. The information illustrated in the user interface may be based on real data and actual event that occurred in the industrial environment [0161].)
As per claims 4 and 13 (Currently Amended), Barak in view of deSa teach the computer-implemented method of claim 1 and the computer system of claim 10, Barak teaches the method further comprising: determining, by the one or more processors, whether the at least one task has been completed; and in response to determining that the at least one task has been completed, displaying, by the one or more processors, at least one record corresponding to the at least one task on the one or more user interfaces of one of the one or more devices. (Barack e.g. In addition, system 100 may also include a control room 140, in which one or more safety-managers, supervisors, managers, and other employees may convene for controlling safety aspects in the industrial environment [0044]. Optionally, the control room screen may present information associated with tasks. For example, details on planned tasks, details on scheduled tasks, details on outstanding tasks, and details on recently completed tasks [0045].)
As per claims 5 and 14 (Original), Barak in view of deSa teach the computer-implemented method of claim 1 and the computer system of claim 10, Barak teaches the method further comprising: assigning, by the one or more processors, the at least one task to the at least one user, the assigning based on the environment information, the task information, and the user information (Barak e.g. Processing device 202 may be configured to generate different first displays to different employees, each including different risk-mitigating instructions selected for a respective recipient employee based on a role of the employee. Processing device 202 may be configured to retrieve from at least one database information pertaining to the different employees, including information pertaining to each of the employees which includes: safety-related historical data pertaining to the respective employee and information indicative of at least one of the respective employee's: health parameters, professional qualifications, and reviews; determine for each of the employees a new action for mitigating the safety-related threat based on the retrieved data associated with the respective employee, wherein different actions are determined for different employees [0155].)
As per claims 6 and 15 (Previously Presented), Barak in view of deSa teach the computer-implemented method of claim 1 and the computer system of claim 10, Barak teaches the method further comprising: analyzing, by the one or more processors, the assessment data to determine at least one equipment recommendation for the at least one task (Barak e.g. System 100 may include at least one sensing device 105 that may (or may not) be associated with employee 110, a server 115 operatively connected to a database 120, and an output unit 125 associated with the industrial environment [0038]. In one example, server 115 may be a cloud server that processes data received from one or more sensing devices (e.g., sensing devices 105A-105E) and processes the data to determine a risk score of a task and/or to identify hazards in the industrial environment [0040]. Server 115 may also process the received data to determine recommendations for preventing accidents [0040]. When the task risk score of the scheduled task is above the first threshold, the processing device may issue a notice prohibiting the execution of the scheduled task (block 508) and provide recommendations to minimize the risk of the scheduled task (block 510) [0084]. Process 500 continues when the processing device provides recommendations for a scheduled task (block 516). In one embodiment, the recommendations for a scheduled task may include checklists, relevant warnings, suggested tools, and more [0085].)
As per claim 7 (Original), Barak in view of deSa teach the computer-implemented method of claim 1, Barak teaches wherein the one or more devices include one or more sensors, one or more mobile devices, one or more internal computing systems, and/or one or more external computing systems (Barak e.g. System 100 for analyzing information collected from an industrial environment may represent a computer-based system that includes computer system components, desktop computers, workstations, tablets, handheld computing devices, memory devices, and/or internal network(s) connecting the components [0037]. System 100 may include at least one sensing device 105 that may (or may not) be associated with employee 110, a server 115 operatively connected to a database 120, and an output unit 125 associated with the industrial environment [0038]. System 100 may analyze data acquired by a plurality of sensing devices 105 to determine a risk score of a task and/or to identify hazards in the industrial environment. Sensing device 105 may include an image capturing device, such as a fixed security camera 105A, autonomous robotic devices with cameras, drones with cameras, etc. In another example, sensing device 105 may include a wireless communication device, such as a worker's handheld communication device 105C, a tablet, a mobile station, a personal digital assistant, a laptop, etc. Sensing device 105 may include a wearable device, such as a smart helmet 105B, smart protective gear, smart glasses, a clip-on camera. In addition, sensing device 105 may be configured to operate manually, remotely, or autonomously [0039].)
As per claim 8 (Original), Barak in view of deSa teach the computer-implemented method of claim 1, Barak teaches wherein the additional data includes additional environment information, additional task information, or additional user information (Barak e.g. Real-time information may be obtained from one or more cameras located in the industrial environment, one or more communication devices of employees in the industrial environment, wearable sensors on employees in the industrial environment, operational technology (OT) sensors, environmental sensors, sensors associated with working tools, and more [0034]. The safety management system described in the present disclosure collects information from a wide variety of sources, processes the diverse information to generate a consolidated database in which safety-related information from different sources (employees, sensors, regulations, protocols, task schedules, work-permits, and so on) is stored in an interconnected fashion [0110].)
As per claim 9 (Previously Presented), Barak in view of deSa teach the computer-implemented method of claim 1, Barak also teaches wherein the one or more thresholds further comprise an environmental threshold or a task threshold wherein the at least one violation further comprises at least one or an environment violation or a task violation (Barak e.g. Embodiments of the present disclosure further include determining a predicted risk score of the scheduled task and determining a change in the risk score of the task. The term "risk score" refers to a score that can be assigned based on comparing synergy data to a risk predictor model. A risk score can have a standard value (e.g., a number) or a multi-value threshold (e.g., a line on a graph). The system may receive real-time information and update the risk score based on events detected using the real-time information. When the actual risk score of a task is above a certain threshold, the system may initiate a remedial action to prevent a work accident. Examples of remedial actions, include transmitting location-based warning messages to employees, displaying the detected hazards on a personalized map, performing an automatic shutdown, and creating customized inspection tour based on the detected locations of the plurality of hazards [0036]. System 100 may analyze data acquired by a plurality of sensing devices 105 to determine a risk score of a task and/or to identify hazards in the industrial environment [0039].)
As per claim 21 (Previously Presented), Barak in view of deSa teach the computer-implemented method of claim 1, Barak in view of deSa teach further comprising in response to determining the health violation: determining that the at least one task was not completed; and automatically updating a database record to indicate equipment to be used to perform a second task, wherein the second task comprises a repeat of the at least one task, wherein the equipment is selected in order to perform the second task without causing a second health violation while performing the second task.
Barak teaches in response to determining a task violation determining that the at least one task was not completed; and automatically updating a database record to indicate equipment to be used to perform a second task, wherein the second task comprises a repeat of the at least one task, wherein the equipment is selected in order to perform the second task without causing a second task violation while performing the second task. (Barack e.g. The methods and systems include receiving from a memory device task scheduling information that includes details of a plurality of tasks associated with the industrial environment, wherein the plurality of tasks includes multiple ongoing tasks which are currently being executed and multiple future tasks which are scheduled to be executed at a later time;...receiving task execution modification information for at least some of the plurality of tasks, including at least three of: (a) detected changes in performances of an employee assigned to the task, (b) detected changes in planned locations of the task, (c) detected changes in tools expected to be used in the task, (d) detected changes in materials expected to be used in the task, (e) detected changes in an expected start time of the task, (f) detected changes in expected duration of the task, and (g) detected changes in an expected weather during the task; [0007]. The first synergy data may include details of at least one handover event expected to happen while the task is taking place. The handover event may be an employee shift change during the task, a material change during the task, a tool change during the task, a supervisor change during the task, and a change from working during daytime and nighttime. For example, the system may detect that during the task two of the workers are expected to be replaced (e.g., it is the end of their shift), this change will increase the risk score of the scheduled task ([0035] and [0066]). The control room screen may present information associated with tasks. For example, details on planned tasks, details on scheduled tasks, details on outstanding tasks, and details on recently completed tasks . In addition, the control room screen may present information associated with safety events, hazards, and/or potentials risks. For example, the control room screen may present details of recent hazards that were reported and treated or not treated yet. Optionally, the control room screen may have a feature of presenting a summary of events for assisting in shift replacement [0045]. Real-time intervention-system 100 may detect that a change to the operating environment or a control indicator is not as expected ( e.g., machine operational status, weather, other tasks, worker specific risk profile) or a combination of factors creates a risk score that is unacceptable [0059]. System 100 may initiate a remedial action, such as triggering real-time alerts, preventing the task from being performed by shutting down connected machines, or making the task paused or locked [0059]. The processing device may retrieve data stored in the employees' database 120 relating to the specific employee (e.g., employee 754B) and retrieve data stored in a task database (e.g., may also be part of database 120) relating to at least one task associated with the specific area of the industrial environment. For example, details about a task being executed or recently completed in building 762C [0094].The at least one task may include an ongoing task which is scheduled for execution before and after a shift change in the industrial environment. In such handover event-and especially when there is a safety risk or ongoing safety event-it is critical that both the current employee assigned to the task and the employee replacing them on the shift change will be coordinated. Such coordination may be facilitated by presenting to each of them coordinated data, which is relevant to their part of the task, before and after the handover [0149]. FIG. 12 illustrates method 1200 for providing real-time safety information at a plurality of locations within an industrial environment [0132]. Step 1240 includes repeatedly updating a common real-time events overview report based on changes in at least one of: the task scheduling information, the real-time sensor information, and the task execution modification information [0136]. FIG. 14 illustrates method 1400 for adapting a safety management system to changing risks, in accordance with examples of the presently disclosed subject matter [0164]. Step 1460 includes updating a safety database to include data pertaining to the new safety-related risk for the at least one object used during the execution of the first task. The safety database may be database 120 [0175]. The task details may be received from a computer, from a database and/or from a user (e.g., manager, supervisor) using a dedicated user interface [0176]. The details of the second task may include, for example, characteristics of the task the include at least one of: an estimated start time of the task, an identity of employees expected to participate in the task, an expected time duration of the task, potential accidents associated with the task, potential accidents associated with the identity of employees, types of materials expected to be used in the task, and types of tools expected to be used in the task [0176].
Barak does not explicitly teach, however, deSa teaches determining a health violation (see claim 1 response).
The Examiner submits that before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify Barak’s real-time safety information system to include determining health violations from health vitals and thresholds as taught by deSa in order to unobtrusively monitor the functional status and safety of patients [and workers] over time for earlier identification of exacerbations or decline beyond clinically important thresholds (deSa e.g. [0045]-[0046]).
As per claim 22 (Currently Amended), Barak in view of deSa the computer-implemented method of claim 1, Barak does not explicitly teach, however, deSa teaches further comprising: in response to determining the health violation: automatically transmitting a first message indicative of the health violation to an external computing system wherein the external computing system is configured to facilitate providing of assistance that is responsive to the health violation to the at least one user and automatically transmitting a second message indicative of the health violation to an internal computing system (deSa e.g., A method comprises: receiving at least one measurement from at least one sensor; determining an activity of a patient based on the received at least one measurement; and generating alerts when trends deviate from a set of pre-defined thresholds [0007]. The example of FIG. 3 shows an embedded, layered system of information and visualization. The dashboard's layered system provides insight into daily routine of living activities, history of specific activities and related anomalies, and vital signs of heart rate (HR) and respiratory rate (RR) as well as vital sign histories. An algorithm creates an optional layered alarm system of user-tailored alerts via the dashboard from low-threshold signals to caregivers and high-threshold alerts to healthcare professionals [0037]. The system has an optional inbuilt crisis alert mechanism that is triggered into a red-yellow-green signal to caregivers and clinical managers [0042]. The dashboard architecture comprises a plurality of sensors 300 (e.g., the sensors 1-5 of FIG. 1) that provide collected information to an edge computing device or module 302 (e.g., a processor) on premises (i.e., at the location, home, etc., of the monitored patient). The edge computing device transmits the collected information to the cloud 30 (FIG. 1) [0043].), Barak teaches wherein the internal computing system is configured to update a start time of a second task to be performed in a location of the at least one user. (Barak e.g. FIG. 12 illustrates method 1200 for providing real-time safety information at a plurality of locations within an industrial environment [0132]. Step 1230 includes receiving task-execution modification-information for at least some of the plurality of tasks. The modification information is any information which is indicative of changes which occurred, or which are planned or expected to occur in the execution of the task. Such information may be received from employees performing the task, from managers, from safety supervisors, from other people, from sensors, from ERP systems, from other data systems and databases, and so on. Some examples of task-execution modification information include one or more of: detected changes in an expected start time of the task;...[0135]. Step 1240 includes repeatedly updating a common real-time events overview report based on changes in at least one of: the task scheduling information, the real-time sensor information, and the task execution modification information [0136]. The common real-time events overview report may be stored in one or more databases, or by any other database available to the safety management system executing method 1200 (e.g., system 100) [0136]. For example, the data collected at the previous steps may indicate that a task which takes place at a lab will continue until 17:00 instead of 16:30. Since another task which is scheduled for 16:45 at the labs involves laser emission, there is an eye safety hazard for the employees performing the earlier task. Such a hazard may be resolved in several ways (e.g., postponing the second task, instructing employees of the earlier task to wear laser safety goggles, etc.) in the following steps [0137].)
The Examiner submits that before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify Barak’s real-time safety information system to include determining and communicating health violations of users as taught by deSa in order to prevent accidents, complications, and more severe exacerbations (deSa e.g. [0045]).
As per claim 24 (Currently Amended), Barak in view of deSa teach the computer-implemented method of claim 1, Barak teaches further comprising: in response to validating the environment information and the user information of the assessment data: generating a predictive preview feature for the at least one task, wherein the predictive preview feature is indicative of one or more operations for performing the at least one task; and causing one or more graphical representations of the predictive preview feature to be rendered on a graphical user interface of a mobile device of the at least one user, wherein the mobile device is one of the one or more devices. (Barak e.g. System 100 may analyze data acquired by a plurality of sensing devices 105 to determine a risk score of a task and/or to identify hazards in the industrial environment. Sensing device 105 may include an image capturing device, such as a fixed security camera 105A, autonomous robotic devices with cameras, drones with cameras, etc. In another example, sensing device 105 may include a wireless communication device, such as a worker's handheld communication device 105C, a tablet, a mobile station, a personal digital assistant, a laptop, etc. [0039]. Sensing device 105 and/or server 115 may communicate with output unit 125 to present information derived from processing data acquired by sensing devices 105 [0042]. For example, output unit 125 may display identified real-time hazards and potential hazards on a personalized map together with visual indicators of the hazard's severity and the hazard's type [0042]. Consistent with the systems and methods described above, it is noted that data collected from employees (as well as other data collected throughout method 1200) may be used for generating visual representation indicative of differences between the planning of tasks to what actually happened during execution [0143]. Optionally, the visual representation of the actual execution of the threatened task may be generated based on the task execution information from the plurality of sensors in the industrial environment and the debriefing-responses of the first employee to the debriefing questionnaire [0143]. In other embodiments, the first display of the common events overview report may be displayed on at least one mobile communications device of the at least one employee and the second display of the common events overview report may be displayed on a control-room computer ([0147] and [0157]).)
Claim 23 is rejected under 35 U.S.C. 103 as being unpatentable over Barak et al. (US 2020/0202471 A1) in view of deSa et al. (US 2021/0057093 A1) and in further view of Saripalli et al. (US 2020/0337648 A1).
As per claim 23 (Currently Amended), Barak in view of deSa and Saripalli teach the computer-implemented method of claim 2, wherein the machine learning model comprises a transformer machine learning network and wherein the computer-implemented method further comprises: training the transformer machine learning network using historical health vitals and historical health acceptable ranges; and updating the transformer machine learning network based on the real-time health vitals.
Barak teaches a real-time safety information system that uses a machine learning model comprising various deep learning techniques to analyze real-time data (Fig. 1, [0034, [0036], and [0067]).
Barak does not explicitly teach, however deSa teaches training a machine learning model using historical health vitals and historical acceptable ranges and updating the machine learning model based on health vitals (deSa e.g. Several standard machine learning methods are applied and non-standard models to develop time sequences 212, feature generation 215, activity classification 220, activity discovery 225, and human subject attribution 230 before the output 235 is developed that is related to routine activities 240, 250 and anomaly detections 232 [0036]. Neural network (specifically Long Short-Term Networks) and deep learning models are continually updated with time series inputs of labeled activities, mobility measures, discovered activities sequences, and vital signs. These models identify significant changes in these inputs over time, allowing the recognition of anomalous activity on the part of the monitored individual ([0040] and [0053]). To train classification algorithms, time-stamped ground truth data is collected at the time of installation based on activation of alternating current electrical devices and water fixtures, plus scripted human activities (which may involve the monitored individual or others). In some embodiments, a monitored individual may wear an accelerometer, gyroscope, and/or radio frequency identification tag to compile additional ground truth model training data (for a period of two weeks or less). These algorithms are distinct due to the nature of the data source that predicts a highly validated behavioral activities of daily living of a human subject along with RR/HR [0041].)
The Examiner submits that before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify Barak’s real-time safety information system’s machine learning process to include training and updating the model using historical and real-time health vitals and acceptable ranges as taught by deSa in order to unobtrusively monitor the functional status and safety of patients [and workers] over time for earlier identification of exacerbations or decline beyond clinically important thresholds (deSa e.g. [0045]-[0046]).
Barak nor deSa explicitly teach the machine-learning model comprises a transformer machine-learning network.
However, Saripalli teaches a machine-learning model that comprises and transformer network (Saripalli e.g. Systems, apparatus, instructions, and methods for medical machine time-series event data processing are disclosed (Abstract). Machine learning can be used to help configure, monitor, and update the medical workflow and devices [0033]. As shown in an example system 100 of FIG. 1, one or more medical devices 110 (e.g., ventilator, anesthesia machine, intravenous (IV) infusion drip, etc.) administer to a patient 120, while one or more monitoring devices 130 (e.g., electrocardiogram (ECG) sensor, blood pressure sensor, respiratory monitor, etc.) gather data regarding patient vitals, patient activity, medical device operation, etc. Such data can be used to train an AI model, can be processed by a trained AI model, etc. [0057]. Other neural networks include transformer networks, graph neural networks, etc. A transformer or transformer network is a neural network architecture that transforms an input sequence to an output sequence using sequence transduction or neural machine translation (e.g., to process speech recognition, text-to-speech transformation, etc.), for example [0071]. In certain examples, a transformer is applied to sequence and time series data [0072]. FIG. 4E shows an example transformer neural network 440 including three input stages and five output stages to transform an input sequence into an output sequence [0073]. FIG. 7 illustrates a schematic of an example system 700 to predict medical machine events using patient waveform data. The example system or apparatus 700 includes patient physiological signal data 710 and medical machine event data 720 provided to an aggregator 730 [0100]. The example classifier 750 of FIG. 7 is used to apply one or more AI models to the data samples of interest 740. For example, the classifier 750 can be configured, selected, triggered, and/or otherwise determined to apply a hybrid RL network model 752, a transformer network model 754, an LSTM network model 756, and/or a GNN model 758 to one or more of the samples of interest 740. The AI model 752-758 applied can depend on a task associated with the request for data processing via the apparatus 700, an input source for the data 710, 720, a target for the classification and/or other predictive output, etc. [0102]. For example, the apparatus 700 can be used to predict one or more future medical machine events and summarize pertinent past medical machine events related to the predicted one or more future medical machine events using a consistent input of time series data related to a patient [0103].)
The Examiner submits that before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify Barak in view of deSa’s real-time safety information system’s machine learning model to include a transformer network as taught by Saripalli because the transformer requires less computation to train and is a much better fit for modern machine learning hardware, speeding up training by up to an order of magnitude (Saripalli e.g. [0072]).
Conclusion
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 Ayanna Minor whose telephone number is (571)272-3605. The examiner can normally be reached M-F 9am-5 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, Jerry O'Connor can be reached at 571-272-6787. 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.
/A.M./Examiner, Art Unit 3624
/Jerry O'Connor/Supervisory Patent Examiner,Group Art Unit 3624
Prosecution Insights