DETAILED ACTION
This Office Action is in response to the Applicants' communication filed on June 8, 2026, which amends the independent claim 1, adds new dependent claims 2-20, and presents arguments, is hereby acknowledged. Claims 1-20 are currently pending and have been examined..
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 .
Response to Amendment
Applicant’s arguments filed on June 8, 2026, have been fully considered.
Applicant argues that by this response, the independent claim 1 is hereby amended to add new limitation “determining a subject-specific metric statistic using the metric features from the meta data; determining whether the metric features from the meta data are associated with a desired metric statistic such that the subject-specific metric statistic is expected to meet the desired metric statistic” in order to overcome the 35 U.S.C. §103 rejection.
Examiner replies that the amended claims with new limitation may overcome the cited portions of the prior arts. However, a newly found art, Rodriguez-Llorente, etc. (US 20140073865 A1) teaches that determining a subject-specific metric statistic using the metric features from the meta data (See Rodriguez-Llorente: Fig. 1, and [0034], “In some embodiments, the processing equipment may generate value pairs each including a first sample point of the physiological signal and a second point of the physiological signal spaced apart by a particular spacing based on a calculated value. The processing equipment may determine a best fit linear relationship based on the plurality of value pairs, determine at least one statistical metric based on the linear relationship and the value pairs, and qualify or disqualify the calculated value based on the at least one statistical metric. In some embodiments, the statistical metric may include a standard error between the value pairs and the linear relationship, as well as a slope of the linear relationship. In some embodiments, qualifying or disqualifying the calculated value may include determining a value indicative of the probability, relative to a predetermined probability distribution function, of the at least one statistical metric being outside of a set of bounding values. In some embodiments, the processing equipment may determine at least one additional spacing other than the particular spacing”; Fig. 17 and [0305], “Step 1710 may include processing equipment determining an algorithm setting based on the area ratios of step 1708. In some embodiments, the processing equipment may analyze a sequence of ratio values to determine a metric. For example, the 25% largest value (e.g., larger than about 75% of ratio values and smaller than about 25% of ratio values) of a sequence of ratio values of positive areas to adjacent negative areas may be compared to a threshold. In a further example, the ratio corresponding to the largest 25% of a sequence of ratio values of positive areas to adjacent negative areas may be compared to a threshold value. In a further example, the average of the largest 25% of a sequence of ratio values of positive areas to adjacent negative areas may be compared to a threshold value. The ratio is expected to be near 1 for data not exhibiting dicrotic notches. For data exhibiting dicrotic notches, the ratios may exhibit two tiers due to the two-tiered shaped of troughs in the difference signal. A first tier will be somewhat close to 1, while the second tier will be significantly larger than 1. By picking the 25% value, the processing equipment will likely pick a value in the middle of second tier for dicrotic notches and therefore be high when dicrotic notches are present. However, when dicrotic notches are not present, the selected value is likely close to 1 because the areas ratios are all generally close to 1 (provided noise is sufficiently low). In some embodiments, the processing equipment may normalize the determined ratios, sort the normalized ratios into a sorted array, and select the value at one fourth of the length of the array of sorted ratios. In some embodiments, the processing equipment may compare the 25% value to the 75% value. For data exhibiting a dicrotic notch the 25% and 75% values should each lie in the middle of the two tiers, while for data not exhibiting a dicrotic notch, the values may be expected to be relatively similar”; and [0370], “In some embodiments, the processing equipment may apply any suitable statistical technique to the two sorted difference signals. For example, the processing equipment may apply a KS Test to the first and second portions by comparing the sorted difference signals to a predetermined distribution. In a further example, the processing equipment may use a function other than a line as a fitting reference. For example, the processing equipment may fit a polynomial of any order to the first and second difference signals, or any other suitable function, and compare the fitted functions to each other or to a reference function”. Note that the statistic metrics such as the linear relationship and value pairs, ratios of area, KS test to the predetermined distribution, etc. are mapped to determining the subject-specific metric statistic); and Ochs, etc. (US 20130289413 A1) teaches that such that the subject-specific metric statistic is expected to meet the desired metric statistic (See Ochs: Fig. 1, and [0015], “Physiological monitoring systems rely on signals that are often corrupted or contain morphologies that produce erratic or unreliable results when processing occurs to extract useful physiological information. The present disclosure describes a system that addresses these concerns by identifying portions of a physiological signal, such as a photoplethysmograph (PPG) signal, that, when processed, are likely to result in erratic or unreliable results”; [0018], “The patient monitoring system may identify the largest contiguous portion of the PPG signal that does not correspond to a large baseline shift and further identify an additional buffer region. The remaining portions of the PPG signal (including the buffer region) may be discarded, replaced, or otherwise ignored. The patient monitoring system may also identify the largest contiguous portion of the PPG signal that does not correspond to an artifact, large pulse-to-pulse variability, or out of range values. The contiguous portion may be required to be of a particular minimum length. The remaining portions of the PPG signal may be discarded, replaced, or otherwise ignored. The remaining usable portion of the PPG signal may be used to determine physiological information such as respiration information”; and Fig. 5, and [0065], “FIG. 5 depicts a flow diagram showing illustrative steps for determining a usable portion of a physiological signal such as a PPG signal in accordance with some embodiments of the present disclosure. Although a number of exemplary steps are disclosed herein, it will be understood that any of the steps depicted in FIG. 5 may be omitted, the order of the steps may be modified, and additional steps may be added. For example, any of the steps for determining signal portions (i.e., steps 502, 504, 506, 508, and 510) may be omitted, and additional signal portion identifying steps may be added. The steps including identifying contiguous signal portions may be performed as depicted in FIG. 5 (i.e., steps 502 and 512), after all signal portions are identified (e.g., at step 512 only), or may be performed individually after each signal portion is identified. Similarly, modifying the determined signal portions may be performed as depicted in FIG. 5 (i.e., at step 512), after some of the signal portions are identified (e.g., at steps 502 and 512), or may be performed individually after each signal portion is identified. In some embodiments different signal modification procedures may be associated with different signal portion determination steps”. Note that extracting features from the PPG signals, using these features to identify the usable portion of the PPG signa, and using these usable portion of PPG signal to determine the physiological information, is mapped to the current claimed limitation of the subject-specific metric statistic is expected to meet the desired metric statistic). The remaining arguments of the applicant are mooted in view of the newly found arts.
Examiner respectfully further replies that the Applicant's arguments have been fully considered and a new ground of rejections have been made. Accordingly, new grounds of rejection are set forth below. Since the new grounds of rejection are necessitated by Applicant's amendments to the claims, the present action is made final.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1, 4, 9-12, and 14-20 are rejected under 35 U.S.C. 103 as being unpatentable over Park, etc. (US 20160120434 A1) in view of Patwardhan, etc. (US 20130060121 A1), further in view of Rodriguez-Llorente, etc. (US 20140073865 A1) and Ochs, etc. (US 20130289413 A1).
Regarding claim 1, Park teaches that a method of qualifying a measurement from a subject (See Park: Fig. 13, and [0170], “FIG. 13 is a schematic illustration of an embodiment of a system and method 3000 for a wearable ECG and/or medical sensor 3002 with transmission capabilities, very similar to the system and/or methods described above in relation to FIGS. 11 and 12. FIG. 13 differs from FIGS. 11 and 12 in that FIG. 13 illustrates alternate sensor channels 3004, 3006 producing alternate outputs and/or extraction 3008 of features 3010. Collection of other channels of data may serve to further augment ECG-extracted features. Data from the alternate sensor channels may be sent whole or specific features 3010 of the data channel may be extracted 3008. In certain embodiments, an alternate data channel may record galvanic skin response/impedance”), the method comprising:
collecting PPG data and meta data from the subject (See Park: Fig. 13, and [0171], “In some embodiments, an alternative data channel may be provided by a pulse oximeter. For example, a photoplethysmogram (PPG) may be generated by the pulse oximeter. The PPG may provide an alternative source for R-peak locations or as a cross-check on R-peak detection by the ECG circuitry. Further, the PPG data channel may be combined with multiple PPG/BioZ channels to output confidence of R-peak detection confidence levels. In further embodiments, SpO2/perfusion via the pulse oximeter may provide further clinical indications of a severe arrhythmia. In certain embodiments, an alternative sensor channel may involve bioimpedance, which may be used to determine heart beat location and/or act as an alternative source for R-peak data. In some embodiments, temperature data may be provided via an alternative sensor channel. This data can be used in conjunction with other metrics of activity to discern activity type, level, and/or sleep. In some embodiments, the alternative data channel may provide information from a clock, for example the time of day or an indication of daytime or nighttime. In certain embodiments, the alternative data channel may be provided by a microphone/stethoscope, providing an audible recording of heart beat. Lastly, an alternative data channel may be provided by a flex or bend sensor which may allow for identification of motion artifacts”; and Fig. 9, and [0148], “In addition, the system may accept other relevant metadata that may help to improve the accuracy of the rhythm analysis, such as user age, gender, indication for monitoring, pre-existing medical conditions, medication information, medical history and the like, and also information on the specific day and time range for each time series submitted to the system”. Note that PPG and metadata are mapped exactly to this claim cited limitations of PPG and metadata);
determining metric features from the meta data (See Park: Figs. 9 and 13, and [0148], “In particular embodiments, a cardiac rhythm inference system 910 may accept additional sources of data, generally described as alternate sensor channels, in addition to R-R interval time series data, to enhance the accuracy and/or value of the inferred results. One additional source of data includes user activity time series data, such as that measured by a 3-axis accelerometer concurrently with the R-R interval time series measurements. In addition, the system may accept other relevant metadata that may help to improve the accuracy of the rhythm analysis, such as user age, gender, indication for monitoring, pre-existing medical conditions, medication information, medical history and the like, and also information on the specific day and time range for each time series submitted to the system. Furthermore, the measurement device might also provide some measure of beat detection confidence, for example, for each R-Peak or for sequential time periods. This confidence measure would be based on analysis the recorded signal that, in typical embodiments, would not be recorded due to storage space and battery energy requirements. Finally, in the particular case that the R-R interval time series data are derived from an ECG signal, the system may accept additional signal features computed from the ECG. These features may include a time series of intra-beat interval measurements (such as the QT or PR interval, or QRS duration), or a time series of signal statistics such as the mean, median, standard deviation or sum of the ECG signal sample values within a given time period”. Note that the metadata related features of R-R, P-R, etc. are mapped to the metric features determined from the metadata);
determining a subject-specific metric statistic using the metric features from the meta data;
determining whether the metric features are associated with (See Park: Figs. 10-11, and [0161], “A wide variety of different types of ECG or comparable biological signal features may be extracted. For example, R-peak locations may be extracted. In certain embodiments, the R-peak locations are extracted via various methods such as: a Pan-Tompkins algorithm (Pan and Tompkins, 1985), providing a real-time QRS complex detection algorithm employing a series of digital filtering steps and adaptive thresholding, or an analog R-peak detection circuit comprising an R-peak detector consisting of a bandpass filter, a comparator circuit, and dynamic gain adjustment to locate R-peaks. The RR-intervals may be calculated from peak locations and used as the primary feature for rhythm discrimination. In embodiments, an R-peak overflow flag may be extracted. If more than a certain number of R-peaks were detected during a given time window such that not all data can be transmitted, a flag may be raised by the firmware. Such an extraction may be used to eliminate noisy segments from analysis, on the basis that extremely short intervals of R-R are not physiologically possible. With similar motivation, an R-peak underflow flag may be extracted to indicate an unrealistically long interval between successive R peaks, provided appropriate considerations for asystole are made in this evaluation. In an alternative implementation with the same goal, the lack of presence of R peaks in a prolonged interval could be associated with a confidence measure, which would describe the likelihood that the interval was clinical or artifact”; and Fig. 18, and [0185], “The Rules Database 14060 may, in some embodiments, store data (for example, instructions, preferences, profile) that establish parameters for the thresholds for analyzing the feature data. In some embodiments, one or more of the databases or data sources may be implemented using a relational database, such as Sybase, Oracle, CodeBase, MySQL, SQLite, and Microsoft® SQL Server, and other types of databases such as, for example, a flat file database, an entity-relationship database, and object-oriented database, NoSQL database, and/or a record-based database”) a desired metric statistic such that the subject-specific metric statistic is expected to meet the desired metric statistic; and
in response to the determining that the metric features from the meta data are associated with (See Park: Fig. 11, and [0167], “The identified arrhythmia locations 1018 are then transmitted 1020 back to the sensor 1002. The transmission 1020 back to the sensor may be accomplished by any communication protocols/technology described herein this section or elsewhere in the specification, for example via Bluetooth. The sensor then reads the transmitted identified locations 1022 and accesses 1024 the areas of memory corresponding to the transmitted identified locations 1022 of the ECG”. Note that the identified location for some events or measurements such as R-R are mapped to the metric features associated with a preset parameter) the desired metric statistic,
categorizing the subject-specific metric statistic as qualified for processing a PPG-based metric for the subject based on the collected PPG data from the subject (See Park: Fig. 11, and [0167], “The identified arrhythmia locations 1018 are then transmitted 1020 back to the sensor 1002. The transmission 1020 back to the sensor may be accomplished by any communication protocols/technology described herein this section or elsewhere in the specification, for example via Bluetooth. The sensor then reads the transmitted identified locations 1022 and accesses 1024 the areas of memory corresponding to the transmitted identified locations 1022 of the ECG. In some embodiments, the sensor applies additional analysis of the identified segments to further build confidence in the arrhythmia identification. This further rhythm confidence determination step 1026 allows for increasing positive predictivity prior to the power-hungry transmission step. In embodiments, if the confidence exceeds a defined threshold the data segment is transmitted. For example, the defined threshold may be a preset value or it may be set per user and monitoring session. In embodiments, the defined threshold may be changed dynamically depending on the nature of the rhythm, the history of accurate detection within the monitoring period, and/or the confidence of the rhythm inference system. Additional analysis may also be performed. Examples of possible analysis techniques include any methods disclosed herein this section or elsewhere in the specification, for example: R-peak amplitude, ECG signal amplitude proxy, ECG signal samples, local ECG signal energy, spectral information, and/or output from a simple machine-learned model”; and [0168], “If the confidence exceeds a threshold as described above, the sensor 1002 may transmit the requested ECG segments 1028 to the processing device via any transmission means described herein this section or elsewhere in the specification. The processing device may complete further analysis on the segments to confirm accuracy of predicted arrhythmia before using data to report to a user and/or physician, as needed”. Note that the arrhythmia is mapped to the category of metric feature, the confidence arrhythmia is mapped to the user-specific metric feature, and threshold exceeded is mapped to it is qualified for further process, and transmitting to the processing unit for additional analysis is mapped to processing the PPG).
However, Park fails to explicitly disclose that determining a subject-specific metric statistic using the metric features from the meta data; and the metric features from the meta data are associated with a desired metric statistic such that the subject-specific metric statistic is expected to meet the desired metric statistic.
However, Patwardhan teaches that the metric features from the meta data are associated with a desired metric statistic (See Patwardhan: Figs. 3-4, and [0081], “In accordance with further aspects of the present technique, the state of the disease progression may be quantified. To that end, a distance metric between the normal statistics and the statistics measured from the test subject, such as the patient 102 is defined. For example, if the normal model is approximated by a Gaussian distribution around the mean values obtained from the normal measurements, then a suitable distance metric may be the Mahalanobis distance”; and [0050], “Additionally, the analysis module 308 is also configured to process the ultrasound images 302, the extracted articular bone surface and/or the segmented joint capsule region to determine an objective volumetric measure of inflammatory changes at the bone joint level. By way of example, the ultrasound images 302 may be processed to provide a direct and intuitive volumetric visualization of the inflamed tissues, volumetric quantification, and/or volumetric Doppler flow assessment. Moreover, the analysis module 308 may also be configured to generate a score or a metric corresponding to the disease state, where the score or metric provides a direct joint-level assessment of the disease state at a point of care, such as a rheumatologist's office. In one embodiment, the analysis module 308 may be configured to determine the score or metric, the disease state and the like by comparing the determined score or metric with measurements corresponding to normal subjects. The normal measurements may be stored in a normal model database 314, for example”. Note that the normal measurements are mapped to the meta data metric features, and the normal statistics of the normal measurements is mapped to the desired metric statistics).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention was effectively filed to modify Park to have the metric features from the meta data are associated with a desired metric statistic as taught by Patwardhan in order to ensures carrying out detection of a disease state in a reliable and inexpensive manner (See Patwardhan: Fig. 4, and [0065], “Once the joint capsule region is segmented, the segmented joint capsule region is analyzed to identify presence of any disease state in the joint, as indicated by step 412. The disease state may include musculoskeletal pathologies such as bone erosion, for example. In accordance with further aspects of the present technique, the segmented joint capsule region is also analyzed to aid in the quantification of the identified disease state. By way of example, the disease state of the segmented joint capsule region may be quantified by providing a score or a metric that is indicative of the severity of the disease. The analysis module 308 is used to identify any disease state, as previously noted with reference to FIG. 3”). Park teaches a method and system that may monitor the subjects by detecting multiple subject physiological states/measurements such as PPG, ECG, etc., using various sensors, extracting features from the PPG and ECG signals, analyzing the features with comparing the features to the thresholds, and feedbacking the analysis results to the users; while Patwardhan teaches a system and method that may identify the disease states of the subjects by comparing the measurement statistics to the normal (desired) statistics to obtain the reliable disease state detection results. Therefore, it is obvious to one of ordinary skill in the art to modify Park by Patwardhan to use the desired metric statistics in the signal feature analysis to obtained the reliable event (disease) detection results. The motivation to modify Park by Patwardhan is “Use of known technique to improve similar devices (methods, or products) in the same way”.
However, Park, modified by Patwardhan, fails to explicitly disclose that determining a subject-specific metric statistic using the metric features from the meta data; and a desired metric statistic such that the subject-specific metric statistic is expected to meet the desired metric statistic.
However, Rodriguez-Llorente teaches that determining a subject-specific metric statistic using the metric features from the meta data (See Rodriguez-Llorente: Fig. 1, and [0034], “In some embodiments, the processing equipment may generate value pairs each including a first sample point of the physiological signal and a second point of the physiological signal spaced apart by a particular spacing based on a calculated value. The processing equipment may determine a best fit linear relationship based on the plurality of value pairs, determine at least one statistical metric based on the linear relationship and the value pairs, and qualify or disqualify the calculated value based on the at least one statistical metric. In some embodiments, the statistical metric may include a standard error between the value pairs and the linear relationship, as well as a slope of the linear relationship. In some embodiments, qualifying or disqualifying the calculated value may include determining a value indicative of the probability, relative to a predetermined probability distribution function, of the at least one statistical metric being outside of a set of bounding values. In some embodiments, the processing equipment may determine at least one additional spacing other than the particular spacing”; Fig. 17 and [0305], “Step 1710 may include processing equipment determining an algorithm setting based on the area ratios of step 1708. In some embodiments, the processing equipment may analyze a sequence of ratio values to determine a metric. For example, the 25% largest value (e.g., larger than about 75% of ratio values and smaller than about 25% of ratio values) of a sequence of ratio values of positive areas to adjacent negative areas may be compared to a threshold. In a further example, the ratio corresponding to the largest 25% of a sequence of ratio values of positive areas to adjacent negative areas may be compared to a threshold value. In a further example, the average of the largest 25% of a sequence of ratio values of positive areas to adjacent negative areas may be compared to a threshold value. The ratio is expected to be near 1 for data not exhibiting dicrotic notches. For data exhibiting dicrotic notches, the ratios may exhibit two tiers due to the two-tiered shaped of troughs in the difference signal. A first tier will be somewhat close to 1, while the second tier will be significantly larger than 1. By picking the 25% value, the processing equipment will likely pick a value in the middle of second tier for dicrotic notches and therefore be high when dicrotic notches are present. However, when dicrotic notches are not present, the selected value is likely close to 1 because the areas ratios are all generally close to 1 (provided noise is sufficiently low). In some embodiments, the processing equipment may normalize the determined ratios, sort the normalized ratios into a sorted array, and select the value at one fourth of the length of the array of sorted ratios. In some embodiments, the processing equipment may compare the 25% value to the 75% value. For data exhibiting a dicrotic notch the 25% and 75% values should each lie in the middle of the two tiers, while for data not exhibiting a dicrotic notch, the values may be expected to be relatively similar”; and [0370], “In some embodiments, the processing equipment may apply any suitable statistical technique to the two sorted difference signals. For example, the processing equipment may apply a KS Test to the first and second portions by comparing the sorted difference signals to a predetermined distribution. In a further example, the processing equipment may use a function other than a line as a fitting reference. For example, the processing equipment may fit a polynomial of any order to the first and second difference signals, or any other suitable function, and compare the fitted functions to each other or to a reference function”. Note that the statistic metrics such as the linear relationship and value pairs, ratios of area, KS test to the predetermined distribution, etc. are mapped to determining the subject-specific metric statistic).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention was effectively filed to modify Park to have determining a subject-specific metric statistic using the metric features from the meta data as taught by Rodriguez-Llorente in order to improve data processing to extract physiological information in presence of noise by effectively limiting the bandwidth of data to be analyzed (See Rodriguez-Llorente: Fig. 10, and [0265], “Physiological monitoring system 10 may use determined algorithm settings to improve data processing (e.g., reduce computational requirements, improve accuracy, reduce the effects of noise) to extract physiological information in the presence of noise. This may be accomplished by effectively limiting the bandwidth of data to be analyzed, performing a rough calculation to estimate a physiological rate or pulse, or otherwise mathematically manipulating physiological data”). Park teaches a method and system that may monitor the subjects by detecting multiple subject physiological states/measurements such as PPG, ECG, etc., using various sensors, extracting features from the PPG and ECG signals, analyzing the features with comparing the features to the thresholds, and feedbacking the analysis results to the users; while Rodriguez-Llorente teaches a system and method that may identify and select the subject-specific PPG data for data processing to extract the metric statistic to reduce the processing data amount to improve efficiency. Therefore, it is obvious to one of ordinary skill in the art to modify Park by Rodriguez-Llorente to determine the subject-specific PPG data for obtaining the metric statistic for the user to reduce bandwidth requirements. The motivation to modify Park by Rodriguez-Llorente is “Use of known technique to improve similar devices (methods, or products) in the same way”.
However, Park, modified by Patwardhan and Rodriguez-Llorente, fails to explicitly disclose that a desired metric statistic such that the subject-specific metric statistic is expected to meet the desired metric statistic.
However, Ochs teaches that a desired metric statistic such that the subject-specific metric statistic is expected to meet the desired metric statistic (See Ochs: Fig. 1, and [0015], “Physiological monitoring systems rely on signals that are often corrupted or contain morphologies that produce erratic or unreliable results when processing occurs to extract useful physiological information. The present disclosure describes a system that addresses these concerns by identifying portions of a physiological signal, such as a photoplethysmograph (PPG) signal, that, when processed, are likely to result in erratic or unreliable results”; [0018], “The patient monitoring system may identify the largest contiguous portion of the PPG signal that does not correspond to a large baseline shift and further identify an additional buffer region. The remaining portions of the PPG signal (including the buffer region) may be discarded, replaced, or otherwise ignored. The patient monitoring system may also identify the largest contiguous portion of the PPG signal that does not correspond to an artifact, large pulse-to-pulse variability, or out of range values. The contiguous portion may be required to be of a particular minimum length. The remaining portions of the PPG signal may be discarded, replaced, or otherwise ignored. The remaining usable portion of the PPG signal may be used to determine physiological information such as respiration information”; and Fig. 5, and [0065], “FIG. 5 depicts a flow diagram showing illustrative steps for determining a usable portion of a physiological signal such as a PPG signal in accordance with some embodiments of the present disclosure. Although a number of exemplary steps are disclosed herein, it will be understood that any of the steps depicted in FIG. 5 may be omitted, the order of the steps may be modified, and additional steps may be added. For example, any of the steps for determining signal portions (i.e., steps 502, 504, 506, 508, and 510) may be omitted, and additional signal portion identifying steps may be added. The steps including identifying contiguous signal portions may be performed as depicted in FIG. 5 (i.e., steps 502 and 512), after all signal portions are identified (e.g., at step 512 only), or may be performed individually after each signal portion is identified. Similarly, modifying the determined signal portions may be performed as depicted in FIG. 5 (i.e., at step 512), after some of the signal portions are identified (e.g., at steps 502 and 512), or may be performed individually after each signal portion is identified. In some embodiments different signal modification procedures may be associated with different signal portion determination steps”. Note that extracting features from the PPG signals, using these features to identify the usable portion of the PPG signa, and using these usable portion of PPG signal to determine the physiological information, is mapped to the current claimed limitation of “a desired metric statistic such that the subject-specific metric statistic is expected to meet the desired metric statistic”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention was effectively filed to modify Park to have the metric features from the meta data are associated with a desired metric statistic as taught by Ochs in order to enable identifying the contiguous portion of the PPG signal that does not include first, second and third portions using the processing equipment, thus determining physiological information (See Ochs: Fig. 1, and [0002], “The method includes identifying a contiguous portion of the PPG signal that does not include the first portion, the second portion, and the third portion, wherein the contiguous portion is at least a minimum length, and determining the physiological information based on the contiguous portion”). Park teaches a method and system that may monitor the subjects by detecting multiple subject physiological states/measurements such as PPG, ECG, etc., using various sensors, extracting features from the PPG and ECG signals, analyzing the features with comparing the features to the thresholds, and feedbacking the analysis results to the users; while Ochs teaches a system and method that may identify the usable portion of the continuous PPG signal of minimum length and meeting the desired metric statistic to determine the subject-specific metric statistic. Therefore, it is obvious to one of ordinary skill in the art to modify Park by Ochs to identify the usable PPG continuous signal portion that meets the desired metric static requirement and use the usable portion PPG signal for analyzing. The motivation to modify Park by Ochs is “Use of known technique to improve similar devices (methods, or products) in the same way”.
Regarding claim 4, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach all the features with respect to claim 1 as outlined above. Further, Park teaches that the method of claim 1, wherein collecting the meta data comprises determining the meta data from sensor data collected via a video imaging device or an audio sensing device (See Park: Fig. 10, and [0165], “Once the feature extraction as described above is completed, various features 1008 may then be transmitted 1010 to a processing device/server 1012. The features 1008 (and alternate sensor channel data and/or features as described below) are transmitted 1010 at regular intervals to a processor 1012 that is not a physical part of the sensor 1002. The interval definition may be pre-set or, configurable with each use, or dynamically configurable. Transmission 1010 of features 1008 may also be bundled and sent when another reason for communication exists, such as transmission of symptomatic data (described in greater detail below in relation to FIG. 16). In certain embodiments, the processing device 1012 may be: a cloud-based server, a physical server at a company location, a physical server at patient or clinic location, a smartphone, tablet, personal computer, smartwatch, automobile console, audio device and/or an alternate device on or off-site. In particular embodiments, the transmission 1010 may utilize short-range RF communication protocols, such as: Bluetooth, ZigBee, WiFi (802.11), Wireless USB, ANT or ANT+, Ultrawideband (UWB), and/or custom protocols. The transmission 1010 may be via infrared communication, such as IrDA and/or inductive coupling communication, such as NFC. In certain embodiments, transmission may be accomplished via cellular data networks and/or wired communication protocols, such as: USB, Serial, TDMA, or other suitable custom means”; and Fig. 13, and [0183], “The exemplary computing device 13000 may include one or more I/O interfaces and devices 13110, for example, a touchpad or touchscreen, but could also include a keyboard, mouse, and printer. In one embodiment, the I/O interfaces and devices 13110 include one or more display devices (such as a touchscreen or monitor) that allow visual presentation of data to a user. More particularly, a display device may provide for the presentation of GUIs, application software data, and multimedia presentations, for example. The computing system 13000 may also include one or more multimedia devices 13140, such as cameras, speakers, video cards, graphics accelerators, and microphones, for example”).
Regarding claim 9, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach all the features with respect to claim 1 as outlined above. Further, Ochs teaches that the method of claim 1, wherein the desired metric statistic comprises an error for the PPG-based metric (See Ochs: Fig. 1, and [0016], “A patient monitoring system may receive a physiological signal such as a PPG signal. The patient monitoring system may identify portions of the PPG signal that correspond to artifacts due to patient motion, invalid measurements, sensor errors (e.g., sensor off or disconnected), or other reasons. These portions of the signal may be identified (e.g., flagged) and may be associated with the PPG signal such that portions of the PPG signal may be identified for additional processing. The patient monitoring system may also determine SpO.sub.2 and pulse rate values, and determine whether these values are out of range”).
Regarding claim 10, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach all the features with respect to claim 9 as outlined above. Further, Ochs teaches that the method of claim 9, wherein the error comprises a mean error or a standard deviation error (See Ochs: Fig. 6, and [0067], “FIG. 6 shows an illustrative PPG signal 602, baseline signal 604, and absolute value signal 606 in accordance with some embodiments of the present disclosure. These exemplary signals may illustrate identifying a large baseline shift in accordance with some embodiments of the present disclosure. PPG signal 602 may be filtered to generate baseline signal 604. Although PPG signal 602 may be filtered in any suitable manner, in an exemplary embodiment PPG signal may be filtered with a 0.07 to 0.7 Hz 3rd order Butterworth filter. To achieve zero phase change, PPG signal 602 may be filtered twice, once in each direction. A threshold 608 may be calculated from the baseline signal. The thresholds may be predetermined or may be dynamic and may be based on, for example, a moving average (weighted or fixed). In an exemplary embodiment threshold 808 may be based on the standard deviation of baseline signal 604 (e.g., a 2.9 times the standard deviation of baseline signal 604)”).
Regarding claim 11, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach all the features with respect to claim 1 as outlined above. Further, Rodriguez-Llorente teaches that the method of claim 1, wherein the PPG-based metric comprises a blood pressure metric (See Rodriguez-Llorente: Fig. 1, and [0179], “The present disclosure is directed towards determining physiological information including physiological rate information. A physiological monitor may determine one or more physiological parameters such as, for example, pulse rate, respiration rate, blood oxygen saturation, blood pressure, or any other suitable parameters, based on one or more signals received from one or more sensors. For example, a physiological monitor may analyze a photoplethysmographic (PPG) signal for oscillometric behavior associated with a pulse rate, a respiration rate, or both. Physiological signals may include desired and undesired signal components. For example, physiological signals may include one or more noise components, which may include the effects of ambient light, electromagnetic radiation from powered devices (e.g., at 50 Hz or 60 Hz), subject movement, any other non-physiological signal component or undesired physiological signal component, or any combination thereof”).
Regarding claim 12, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach all the features with respect to claim 11 as outlined above. Further, Rodriguez-Llorente teaches that the method of claim 11, wherein the desired metric statistic comprises an error for the blood pressure metric (See Rodriguez-Llorente: Figs. 111-112, and [0640], “Step 11118 may include the processing equipment analyzing the first sorted difference signal, the second sorted difference signal, and the third sorted difference signal to determine one or more first metrics. In some embodiments, the analysis may include determining and comparing shape metrics for the sorted difference signals. Shape metrics may include a best fit slope, end points of a sorted difference signal, length of a sorted difference signal, any other suitable shape or geometrical metric, or any combination thereof. In some embodiments, the processing equipment may perform a KS test, comparing the sorted difference signal to a predetermined function. In some embodiments, the analysis may include determining a standard error between any two sorted difference signals of the three sorted difference signals. For example, FIG. 112 is a panel of illustrative plots showing sorted difference signals for a lag, half lag, and double lag segment of physiological data, in accordance with some embodiments of the present disclosure. Plots 11200, 11210, and 11220 show three sorted difference signals, corresponding to one lag, one half lag, and double lag segments (e.g., where the lag is a calculated value), respectively, for which the lag is indicative of a period of a physiological rate. Plots 11230, 11240, and 11250 show three sorted difference signals, corresponding to one lag, one half lag, and double lag segments (e.g., where the lag is a calculated value), respectively, for which the lag is indicative of double a period of a physiological rate. Plots 11260, 11270, and 11280 show three sorted difference signals, corresponding to one lag, one half lag, and double lag segments (e.g., where the lag is a calculated value), respectively, for which the lag is indicative of one half of a period of a physiological rate. Accordingly, the processing equipment may distinguish between conditions when a calculated lag value is a harmonic of the lag value associated with a physiological rate. For example, when the correct lag value is calculated, the lag and double lag sorted difference signals are similar in shape, while the half lag sorted difference signal has a different profile because it includes only a portion of a period of physiological data. In a further example, when double the correct lag value is calculated, the lag, double lag, and half lag sorted difference signals are all similar in shape, because all include at least one full period of physiological data. In a further example, when one half of the correct lag value is calculated, the lag, double lag, and half lag sorted difference signals are all different in shape, because they include a half period, a full period, and a quarter period, respectively of physiological data”).
Regarding claim 14, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach all the features with respect to claim 1 as outlined above. Further, Ochs teaches that the method of claim 1, further comprising, in response to categorizing the subject-specific metric statistic as qualified (See Ochs: Fig. 3, and [0055], “Another example of operations performed by pre-processor 312 may be to identify one or more portions of an analysis window of data (e.g., a 45-second window of the 9 previous 5-second sampling windows) that may include invalid or questionable data. Although such portions of the data may be identified in any suitable manner, in an exemplary embodiment the portions of data may be identified based on one or more flags, PPG data, pulse rate values, or SpO.sub.2 values as described herein. Pre-processor 312 may also modify the PPG signal based on the identified portions invalid or questionable data to generate a usable portion of the PPG signal. Although the PPG signal may be modified in any suitable manner, in an exemplary embodiment any samples of PPG data corresponding to invalid or questionable data may be removed from the analysis window of data, attenuated, or replaced with substitute data as described herein”), processing the collected PPG data to determine the PPG-based metric for the subject (See Ochs: Figs. 5-6, and [0068], “] Absolute value signal 606 may be the generated based on the absolute value of baseline signal 604, and compared to threshold 608. Any locations where absolute value signal 606 crosses threshold 608 may be identified (e.g., crossing points 610 and 612). A longest contiguous signal portion 614 between crossing points may be identified. In some embodiments, longest contiguous signal portion 614 may be selected as the usable portion of PPG signal 602. In another embodiment, buffer regions 616 and 618 may be established from crossing points 610 and 612. Buffer regions 616 and 618 may ensure that the sections of data identified as the usable portion of PPG signal 602 are sufficiently far enough away from the portions of the signal that exhibit a large baseline shift. If buffer regions 616 and 618 are implemented, the excluded portions of PPG signal 602 may be the portions associated with signal portions 622 and 624, while the longest contiguous portion of PPG signal 602 may be signal portion 620. In an exemplary embodiment the portions of PPG signal 602 associated with signal portions 622 and 624 may be discarded. In other embodiments the portions of PPG signal 602 associated with signal portions 622 and 624 may be modified, set to zero, attenuated, replaced with white noise, or replaced with cyclical padding, as described herein”. Note that only the usable continuous portion of PPG is buffer and used to extract the features).
Regarding claim 15, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach all the features with respect to claim 1 as outlined above. Further, Ochs teaches that the method of claim 1, further comprising omitting processing of PPG features from the collected PPG data in response to not categorizing the subject-specific metric statistic as qualified (See Ochs: Figs. 5-6, and [0068], “] Absolute value signal 606 may be the generated based on the absolute value of baseline signal 604, and compared to threshold 608. Any locations where absolute value signal 606 crosses threshold 608 may be identified (e.g., crossing points 610 and 612). A longest contiguous signal portion 614 between crossing points may be identified. In some embodiments, longest contiguous signal portion 614 may be selected as the usable portion of PPG signal 602. In another embodiment, buffer regions 616 and 618 may be established from crossing points 610 and 612. Buffer regions 616 and 618 may ensure that the sections of data identified as the usable portion of PPG signal 602 are sufficiently far enough away from the portions of the signal that exhibit a large baseline shift. If buffer regions 616 and 618 are implemented, the excluded portions of PPG signal 602 may be the portions associated with signal portions 622 and 624, while the longest contiguous portion of PPG signal 602 may be signal portion 620. In an exemplary embodiment the portions of PPG signal 602 associated with signal portions 622 and 624 may be discarded. In other embodiments the portions of PPG signal 602 associated with signal portions 622 and 624 may be modified, set to zero, attenuated, replaced with white noise, or replaced with cyclical padding, as described herein”. Note that not usable portion of the PPG signal is discarded).
Regarding claim 16, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach all the features with respect to claim 1 as outlined above. Further, Park teaches that the method of claim 1, wherein the meta data is received from the subject, from a third party, or from sensor data (See Park: Fig. 1, and [0110], “Another optional data channel that may be added to physiological monitoring device 100 is a channel for detecting flex and/or bend of device 100. In various embodiments, for example, device 100 may include a strain gauge, piezoelectric sensor or optical sensor to detect motion artifact in device 100 itself and thus help to distinguish between motion artifact and cardiac rhythm data. Yet another optional data channel for device 100 may be a channel for detecting heart rate. For example, a pulse oximeter, microphone or stethoscope may provide heart rate information. Redundant heart rate data may facilitate discrimination of ECG signals from artifact. This is particularly useful in cases where arrhythmia such as Supraventricular Tachycardia is interrupted by artifact, and decisions must be made whether the episode was actually multiple shorter episodes or one sustained episode. Another data channel may be included for detecting ambient electrical noise. For example, device 100 may include an antenna for picking up electromagnetic interference. Detection of electromagnetic interference may facilitate discrimination of electrical noise from real ECG signals. Any of the above-described data channels may be stored to support future noise discrimination or applied for immediate determination of clinical validity in real-time”; and Fig. 9, and [0144], “The R-R interval time series 902 data may be extracted from or received from a dedicated heart rate monitor such as a heart rate chest strap or heart rate watch, or a wearable health or fitness device 906, 908 that incorporates heart rate sensing functionality. Alternatively, the R-R interval time series 902 may be derived from a wearable patch designed to measure an ECG signal 904 (for instance, by locating the R peaks in the ECG using a QRS detection algorithm). Furthermore, the R-R interval time series 902 may be estimated from an alternative physiological signal such as that obtained from photoplethysmography (PPG). In this scenario, the peak-to-peak interval time series determined from the PPG signal may be used as an accurate estimate of the R-R interval time series”. Note that the sensors capturing the signals, and extracting physiological features from the sensed signal, is mapped to the metadata from the sensors).
Regarding claim 17, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach all the features with respect to claim 1 as outlined above. Further, Park teaches that the method of claim 1, wherein the PPG data and the meta data are collected via a sensor system comprising a PPG sensor (See Park: Fig. 13, and [0171], “In some embodiments, an alternative data channel may be provided by a pulse oximeter. For example, a photoplethysmogram (PPG) may be generated by the pulse oximeter. The PPG may provide an alternative source for R-peak locations or as a cross-check on R-peak detection by the ECG circuitry. Further, the PPG data channel may be combined with multiple PPG/BioZ channels to output confidence of R-peak detection confidence levels. In further embodiments, SpO2/perfusion via the pulse oximeter may provide further clinical indications of a severe arrhythmia. In certain embodiments, an alternative sensor channel may involve bioimpedance, which may be used to determine heart beat location and/or act as an alternative source for R-peak data. In some embodiments, temperature data may be provided via an alternative sensor channel. This data can be used in conjunction with other metrics of activity to discern activity type, level, and/or sleep. In some embodiments, the alternative data channel may provide information from a clock, for example the time of day or an indication of daytime or nighttime. In certain embodiments, the alternative data channel may be provided by a microphone/stethoscope, providing an audible recording of heart beat. Lastly, an alternative data channel may be provided by a flex or bend sensor which may allow for identification of motion artifacts”).
Regarding claim 18, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach all the features with respect to claim 1 as outlined above. Further, Park teaches that the method of claim 1, further comprising displaying the subject-specific metric statistic via a display of a mobile communication device (See Park: Fig. 1, and [0083], “Additionally, the medical treatment process to actually obtain a cardiac rhythm monitoring device and initiate monitoring is typically very complicated. There are usually numerous steps involved in ordering, tracking, monitoring, retrieving, and analyzing the data from such a monitoring device. In most cases, cardiac monitoring devices used today are ordered by a cardiologist or a cardiac electrophysiologist (EP), rather than the patient's primary care physician (PCP). This is of significance since the PCP is often the first physician to see the patient and determine that the patient's symptoms could be due to an arrhythmia. After the patient sees the PCP, the PCP will make an appointment for the patient to see a cardiologist or an EP. This appointment is usually several weeks from the initial visit with the PCP, which in itself leads to a delay in making a potential diagnosis as well as increases the likelihood that an arrhythmia episode will occur and go undiagnosed. When the patient finally sees the cardiologist or EP, a cardiac rhythm monitoring device will usually be ordered. The monitoring period can last 24 to 48 hours (Holter monitor) or up to a month (cardiac event monitor or mobile telemetry device). Once the monitoring has been completed, the patient typically must return the device to the clinic, which itself can be an inconvenience. After the data has been processed by the monitoring company or by a technician on-site at a hospital or office, a report will finally be sent to the cardiologist or EP for analysis. This complex process results in fewer patients receiving cardiac rhythm monitoring than would ideally receive it”).
Regarding claim 19, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach all the features with respect to claim 1 as outlined above. Further, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach that a system for qualifying a measurement from a subject, the system (See Park: Fig. 13, and [0170], “FIG. 13 is a schematic illustration of an embodiment of a system and method 3000 for a wearable ECG and/or medical sensor 3002 with transmission capabilities, very similar to the system and/or methods described above in relation to FIGS. 11 and 12. FIG. 13 differs from FIGS. 11 and 12 in that FIG. 13 illustrates alternate sensor channels 3004, 3006 producing alternate outputs and/or extraction 3008 of features 3010. Collection of other channels of data may serve to further augment ECG-extracted features. Data from the alternate sensor channels may be sent whole or specific features 3010 of the data channel may be extracted 3008. In certain embodiments, an alternate data channel may record galvanic skin response/impedance”) comprising:
a PPG sensor configured to collect PPG data from the subject (See Park: Fig. 13, and [0171], “In some embodiments, an alternative data channel may be provided by a pulse oximeter. For example, a photoplethysmogram (PPG) may be generated by the pulse oximeter. The PPG may provide an alternative source for R-peak locations or as a cross-check on R-peak detection by the ECG circuitry. Further, the PPG data channel may be combined with multiple PPG/BioZ channels to output confidence of R-peak detection confidence levels. In further embodiments, SpO2/perfusion via the pulse oximeter may provide further clinical indications of a severe arrhythmia. In certain embodiments, an alternative sensor channel may involve bioimpedance, which may be used to determine heart beat location and/or act as an alternative source for R-peak data. In some embodiments, temperature data may be provided via an alternative sensor channel. This data can be used in conjunction with other metrics of activity to discern activity type, level, and/or sleep. In some embodiments, the alternative data channel may provide information from a clock, for example the time of day or an indication of daytime or nighttime. In certain embodiments, the alternative data channel may be provided by a microphone/stethoscope, providing an audible recording of heart beat. Lastly, an alternative data channel may be provided by a flex or bend sensor which may allow for identification of motion artifacts”; and Fig. 9, and [0148], “In addition, the system may accept other relevant metadata that may help to improve the accuracy of the rhythm analysis, such as user age, gender, indication for monitoring, pre-existing medical conditions, medication information, medical history and the like, and also information on the specific day and time range for each time series submitted to the system”. Note that PPG and metadata are mapped exactly to this claim cited limitations of PPG and metadata);
a meta data input device configured to receive meta data from the subject (See Park: Fig. 9, and [0148], “In addition, the system may accept other relevant metadata that may help to improve the accuracy of the rhythm analysis, such as user age, gender, indication for monitoring, pre-existing medical conditions, medication information, medical history and the like, and also information on the specific day and time range for each time series submitted to the system”. Note that PPG and metadata are mapped exactly to this claim cited limitations of PPG and metadata); and
at least one processor in communication with the PPG sensor and the meta data input device, the at least one processor (See Park: Fig. 18, and [0058], “In certain embodiments, a system for assessing physiological sensor data from a patient monitoring device comprises: a computer processor and non-transitory computer-readable media combined with the computer processor configured to provide a program that includes a set of instructions stored on a first server, the set of instructions being executable by the computer processor, and further configured to execute a sensor data inference module of the program; the sensor data inference module of the program storing instructions to: receive physiological sensor data generated by a patient monitoring device, the physiological sensor data associated with a first patient; analyze the physiological sensor data to determine whether one or more points in the physiological data that are likely indicative of one or more predetermined set of conditions; and after determining that at least one of the one or more points in the physiological data is likely indicative of at least one of the one or more predetermined set of conditions, generating an electronic data package for transmission to the patient monitoring device, the electronic data package including location data regarding the at least one of the one or more points in the physiological sensor data that are likely indicative of the at least one of the one or more predetermined set of conditions”) configured to:
collect the PPG data and the meta data from the subject (See Park: Fig. 13, and [0171], “In some embodiments, an alternative data channel may be provided by a pulse oximeter. For example, a photoplethysmogram (PPG) may be generated by the pulse oximeter. The PPG may provide an alternative source for R-peak locations or as a cross-check on R-peak detection by the ECG circuitry. Further, the PPG data channel may be combined with multiple PPG/BioZ channels to output confidence of R-peak detection confidence levels. In further embodiments, SpO2/perfusion via the pulse oximeter may provide further clinical indications of a severe arrhythmia. In certain embodiments, an alternative sensor channel may involve bioimpedance, which may be used to determine heart beat location and/or act as an alternative source for R-peak data. In some embodiments, temperature data may be provided via an alternative sensor channel. This data can be used in conjunction with other metrics of activity to discern activity type, level, and/or sleep. In some embodiments, the alternative data channel may provide information from a clock, for example the time of day or an indication of daytime or nighttime. In certain embodiments, the alternative data channel may be provided by a microphone/stethoscope, providing an audible recording of heart beat. Lastly, an alternative data channel may be provided by a flex or bend sensor which may allow for identification of motion artifacts”; and Fig. 9, and [0148], “In addition, the system may accept other relevant metadata that may help to improve the accuracy of the rhythm analysis, such as user age, gender, indication for monitoring, pre-existing medical conditions, medication information, medical history and the like, and also information on the specific day and time range for each time series submitted to the system”. Note that PPG and metadata are mapped exactly to this claim cited limitations of PPG and metadata),
determine metric features from the meta data (See Park: Figs. 9 and 13, and [0148], “In particular embodiments, a cardiac rhythm inference system 910 may accept additional sources of data, generally described as alternate sensor channels, in addition to R-R interval time series data, to enhance the accuracy and/or value of the inferred results. One additional source of data includes user activity time series data, such as that measured by a 3-axis accelerometer concurrently with the R-R interval time series measurements. In addition, the system may accept other relevant metadata that may help to improve the accuracy of the rhythm analysis, such as user age, gender, indication for monitoring, pre-existing medical conditions, medication information, medical history and the like, and also information on the specific day and time range for each time series submitted to the system. Furthermore, the measurement device might also provide some measure of beat detection confidence, for example, for each R-Peak or for sequential time periods. This confidence measure would be based on analysis the recorded signal that, in typical embodiments, would not be recorded due to storage space and battery energy requirements. Finally, in the particular case that the R-R interval time series data are derived from an ECG signal, the system may accept additional signal features computed from the ECG. These features may include a time series of intra-beat interval measurements (such as the QT or PR interval, or QRS duration), or a time series of signal statistics such as the mean, median, standard deviation or sum of the ECG signal sample values within a given time period”. Note that the metadata related features of R-R, P-R, etc. are mapped to the metric features determined from the metadata),
determine a subject-specific metric statistic using the metric features from the meta data (See Rodriguez-Llorente: Fig. 1, and [0034], “In some embodiments, the processing equipment may generate value pairs each including a first sample point of the physiological signal and a second point of the physiological signal spaced apart by a particular spacing based on a calculated value. The processing equipment may determine a best fit linear relationship based on the plurality of value pairs, determine at least one statistical metric based on the linear relationship and the value pairs, and qualify or disqualify the calculated value based on the at least one statistical metric. In some embodiments, the statistical metric may include a standard error between the value pairs and the linear relationship, as well as a slope of the linear relationship. In some embodiments, qualifying or disqualifying the calculated value may include determining a value indicative of the probability, relative to a predetermined probability distribution function, of the at least one statistical metric being outside of a set of bounding values. In some embodiments, the processing equipment may determine at least one additional spacing other than the particular spacing”; Fig. 17 and [0305], “Step 1710 may include processing equipment determining an algorithm setting based on the area ratios of step 1708. In some embodiments, the processing equipment may analyze a sequence of ratio values to determine a metric. For example, the 25% largest value (e.g., larger than about 75% of ratio values and smaller than about 25% of ratio values) of a sequence of ratio values of positive areas to adjacent negative areas may be compared to a threshold. In a further example, the ratio corresponding to the largest 25% of a sequence of ratio values of positive areas to adjacent negative areas may be compared to a threshold value. In a further example, the average of the largest 25% of a sequence of ratio values of positive areas to adjacent negative areas may be compared to a threshold value. The ratio is expected to be near 1 for data not exhibiting dicrotic notches. For data exhibiting dicrotic notches, the ratios may exhibit two tiers due to the two-tiered shaped of troughs in the difference signal. A first tier will be somewhat close to 1, while the second tier will be significantly larger than 1. By picking the 25% value, the processing equipment will likely pick a value in the middle of second tier for dicrotic notches and therefore be high when dicrotic notches are present. However, when dicrotic notches are not present, the selected value is likely close to 1 because the areas ratios are all generally close to 1 (provided noise is sufficiently low). In some embodiments, the processing equipment may normalize the determined ratios, sort the normalized ratios into a sorted array, and select the value at one fourth of the length of the array of sorted ratios. In some embodiments, the processing equipment may compare the 25% value to the 75% value. For data exhibiting a dicrotic notch the 25% and 75% values should each lie in the middle of the two tiers, while for data not exhibiting a dicrotic notch, the values may be expected to be relatively similar”; and [0370], “In some embodiments, the processing equipment may apply any suitable statistical technique to the two sorted difference signals. For example, the processing equipment may apply a KS Test to the first and second portions by comparing the sorted difference signals to a predetermined distribution. In a further example, the processing equipment may use a function other than a line as a fitting reference. For example, the processing equipment may fit a polynomial of any order to the first and second difference signals, or any other suitable function, and compare the fitted functions to each other or to a reference function”. Note that the statistic metrics such as the linear relationship and value pairs, ratios of area, KS test to the predetermined distribution, etc. are mapped to determining the subject-specific metric statistic),
determine whether the metric features from the meta data are associated with (See Park: Figs. 10-11, and [0161], “A wide variety of different types of ECG or comparable biological signal features may be extracted. For example, R-peak locations may be extracted. In certain embodiments, the R-peak locations are extracted via various methods such as: a Pan-Tompkins algorithm (Pan and Tompkins, 1985), providing a real-time QRS complex detection algorithm employing a series of digital filtering steps and adaptive thresholding, or an analog R-peak detection circuit comprising an R-peak detector consisting of a bandpass filter, a comparator circuit, and dynamic gain adjustment to locate R-peaks. The RR-intervals may be calculated from peak locations and used as the primary feature for rhythm discrimination. In embodiments, an R-peak overflow flag may be extracted. If more than a certain number of R-peaks were detected during a given time window such that not all data can be transmitted, a flag may be raised by the firmware. Such an extraction may be used to eliminate noisy segments from analysis, on the basis that extremely short intervals of R-R are not physiologically possible. With similar motivation, an R-peak underflow flag may be extracted to indicate an unrealistically long interval between successive R peaks, provided appropriate considerations for asystole are made in this evaluation. In an alternative implementation with the same goal, the lack of presence of R peaks in a prolonged interval could be associated with a confidence measure, which would describe the likelihood that the interval was clinical or artifact”; and Fig. 18, and [0185], “The Rules Database 14060 may, in some embodiments, store data (for example, instructions, preferences, profile) that establish parameters for the thresholds for analyzing the feature data. In some embodiments, one or more of the databases or data sources may be implemented using a relational database, such as Sybase, Oracle, CodeBase, MySQL, SQLite, and Microsoft® SQL Server, and other types of databases such as, for example, a flat file database, an entity-relationship database, and object-oriented database, NoSQL database, and/or a record-based database”) a desired metric statistic(See Patwardhan: Figs. 3-4, and [0081], “In accordance with further aspects of the present technique, the state of the disease progression may be quantified. To that end, a distance metric between the normal statistics and the statistics measured from the test subject, such as the patient 102 is defined. For example, if the normal model is approximated by a Gaussian distribution around the mean values obtained from the normal measurements, then a suitable distance metric may be the Mahalanobis distance”; and [0050], “Additionally, the analysis module 308 is also configured to process the ultrasound images 302, the extracted articular bone surface and/or the segmented joint capsule region to determine an objective volumetric measure of inflammatory changes at the bone joint level. By way of example, the ultrasound images 302 may be processed to provide a direct and intuitive volumetric visualization of the inflamed tissues, volumetric quantification, and/or volumetric Doppler flow assessment. Moreover, the analysis module 308 may also be configured to generate a score or a metric corresponding to the disease state, where the score or metric provides a direct joint-level assessment of the disease state at a point of care, such as a rheumatologist's office. In one embodiment, the analysis module 308 may be configured to determine the score or metric, the disease state and the like by comparing the determined score or metric with measurements corresponding to normal subjects. The normal measurements may be stored in a normal model database 314, for example”. Note that the normal measurements are mapped to the meta data metric features, and the normal statistics of the normal measurements is mapped to the desired metric statistics) such that the subject-specific metric statistic is expected to meet the desired metric statistic (See Ochs: Fig. 1, and [0015], “Physiological monitoring systems rely on signals that are often corrupted or contain morphologies that produce erratic or unreliable results when processing occurs to extract useful physiological information. The present disclosure describes a system that addresses these concerns by identifying portions of a physiological signal, such as a photoplethysmograph (PPG) signal, that, when processed, are likely to result in erratic or unreliable results”; [0018], “The patient monitoring system may identify the largest contiguous portion of the PPG signal that does not correspond to a large baseline shift and further identify an additional buffer region. The remaining portions of the PPG signal (including the buffer region) may be discarded, replaced, or otherwise ignored. The patient monitoring system may also identify the largest contiguous portion of the PPG signal that does not correspond to an artifact, large pulse-to-pulse variability, or out of range values. The contiguous portion may be required to be of a particular minimum length. The remaining portions of the PPG signal may be discarded, replaced, or otherwise ignored. The remaining usable portion of the PPG signal may be used to determine physiological information such as respiration information”; and Fig. 5, and [0065], “FIG. 5 depicts a flow diagram showing illustrative steps for determining a usable portion of a physiological signal such as a PPG signal in accordance with some embodiments of the present disclosure. Although a number of exemplary steps are disclosed herein, it will be understood that any of the steps depicted in FIG. 5 may be omitted, the order of the steps may be modified, and additional steps may be added. For example, any of the steps for determining signal portions (i.e., steps 502, 504, 506, 508, and 510) may be omitted, and additional signal portion identifying steps may be added. The steps including identifying contiguous signal portions may be performed as depicted in FIG. 5 (i.e., steps 502 and 512), after all signal portions are identified (e.g., at step 512 only), or may be performed individually after each signal portion is identified. Similarly, modifying the determined signal portions may be performed as depicted in FIG. 5 (i.e., at step 512), after some of the signal portions are identified (e.g., at steps 502 and 512), or may be performed individually after each signal portion is identified. In some embodiments different signal modification procedures may be associated with different signal portion determination steps”. Note that extracting features from the PPG signals, using these features to identify the usable portion of the PPG signa, and using these usable portion of PPG signal to determine the physiological information, is mapped to the current claimed limitation of “a desired metric statistic such that the subject-specific metric statistic is expected to meet the desired metric statistic”), and
in response to determining that the metric features from the meta data are associated with (See Park: Fig. 11, and [0167], “The identified arrhythmia locations 1018 are then transmitted 1020 back to the sensor 1002. The transmission 1020 back to the sensor may be accomplished by any communication protocols/technology described herein this section or elsewhere in the specification, for example via Bluetooth. The sensor then reads the transmitted identified locations 1022 and accesses 1024 the areas of memory corresponding to the transmitted identified locations 1022 of the ECG”. Note that the identified location for some events or measurements such as R-R are mapped to the metric features associated with a preset parameter) the desired metric statistic (See Patwardhan: Figs. 3-4, and [0081], “In accordance with further aspects of the present technique, the state of the disease progression may be quantified. To that end, a distance metric between the normal statistics and the statistics measured from the test subject, such as the patient 102 is defined. For example, if the normal model is approximated by a Gaussian distribution around the mean values obtained from the normal measurements, then a suitable distance metric may be the Mahalanobis distance”; and [0050], “Additionally, the analysis module 308 is also configured to process the ultrasound images 302, the extracted articular bone surface and/or the segmented joint capsule region to determine an objective volumetric measure of inflammatory changes at the bone joint level. By way of example, the ultrasound images 302 may be processed to provide a direct and intuitive volumetric visualization of the inflamed tissues, volumetric quantification, and/or volumetric Doppler flow assessment. Moreover, the analysis module 308 may also be configured to generate a score or a metric corresponding to the disease state, where the score or metric provides a direct joint-level assessment of the disease state at a point of care, such as a rheumatologist's office. In one embodiment, the analysis module 308 may be configured to determine the score or metric, the disease state and the like by comparing the determined score or metric with measurements corresponding to normal subjects. The normal measurements may be stored in a normal model database 314, for example”. Note that the normal measurements are mapped to the meta data metric features, and the normal statistics of the normal measurements is mapped to the desired metric statistics),
categorize the subject-specific metric statistic as qualified for processing a PPG-based metric for the subject based on the collected PPG data from the subject (See Park: Fig. 11, and [0167], “The identified arrhythmia locations 1018 are then transmitted 1020 back to the sensor 1002. The transmission 1020 back to the sensor may be accomplished by any communication protocols/technology described herein this section or elsewhere in the specification, for example via Bluetooth. The sensor then reads the transmitted identified locations 1022 and accesses 1024 the areas of memory corresponding to the transmitted identified locations 1022 of the ECG. In some embodiments, the sensor applies additional analysis of the identified segments to further build confidence in the arrhythmia identification. This further rhythm confidence determination step 1026 allows for increasing positive predictivity prior to the power-hungry transmission step. In embodiments, if the confidence exceeds a defined threshold the data segment is transmitted. For example, the defined threshold may be a preset value or it may be set per user and monitoring session. In embodiments, the defined threshold may be changed dynamically depending on the nature of the rhythm, the history of accurate detection within the monitoring period, and/or the confidence of the rhythm inference system. Additional analysis may also be performed. Examples of possible analysis techniques include any methods disclosed herein this section or elsewhere in the specification, for example: R-peak amplitude, ECG signal amplitude proxy, ECG signal samples, local ECG signal energy, spectral information, and/or output from a simple machine-learned model”; and [0168], “If the confidence exceeds a threshold as described above, the sensor 1002 may transmit the requested ECG segments 1028 to the processing device via any transmission means described herein this section or elsewhere in the specification. The processing device may complete further analysis on the segments to confirm accuracy of predicted arrhythmia before using data to report to a user and/or physician, as needed”. Note that the arrhythmia is mapped to the category of metric feature, the confidence arrhythmia is mapped to the user-specific metric feature, and threshold exceeded is mapped to it is qualified for further process, and transmitting to the processing unit for additional analysis is mapped to processing the PPG).
Regarding claim 20, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach all the features with respect to claim 1 as outlined above. Further, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach that s non-transitory computer-readable medium comprising instructions that, when executed by one or more processors, cause the one or more processors to perform operations (See Park: Fig. 13, and [0170], “FIG. 13 is a schematic illustration of an embodiment of a system and method 3000 for a wearable ECG and/or medical sensor 3002 with transmission capabilities, very similar to the system and/or methods described above in relation to FIGS. 11 and 12. FIG. 13 differs from FIGS. 11 and 12 in that FIG. 13 illustrates alternate sensor channels 3004, 3006 producing alternate outputs and/or extraction 3008 of features 3010. Collection of other channels of data may serve to further augment ECG-extracted features. Data from the alternate sensor channels may be sent whole or specific features 3010 of the data channel may be extracted 3008. In certain embodiments, an alternate data channel may record galvanic skin response/impedance”)compnsmg:
collecting PPG data and meta data from a subject (See Park: Fig. 13, and [0171], “In some embodiments, an alternative data channel may be provided by a pulse oximeter. For example, a photoplethysmogram (PPG) may be generated by the pulse oximeter. The PPG may provide an alternative source for R-peak locations or as a cross-check on R-peak detection by the ECG circuitry. Further, the PPG data channel may be combined with multiple PPG/BioZ channels to output confidence of R-peak detection confidence levels. In further embodiments, SpO2/perfusion via the pulse oximeter may provide further clinical indications of a severe arrhythmia. In certain embodiments, an alternative sensor channel may involve bioimpedance, which may be used to determine heart beat location and/or act as an alternative source for R-peak data. In some embodiments, temperature data may be provided via an alternative sensor channel. This data can be used in conjunction with other metrics of activity to discern activity type, level, and/or sleep. In some embodiments, the alternative data channel may provide information from a clock, for example the time of day or an indication of daytime or nighttime. In certain embodiments, the alternative data channel may be provided by a microphone/stethoscope, providing an audible recording of heart beat. Lastly, an alternative data channel may be provided by a flex or bend sensor which may allow for identification of motion artifacts”; and Fig. 9, and [0148], “In addition, the system may accept other relevant metadata that may help to improve the accuracy of the rhythm analysis, such as user age, gender, indication for monitoring, pre-existing medical conditions, medication information, medical history and the like, and also information on the specific day and time range for each time series submitted to the system”. Note that PPG and metadata are mapped exactly to this claim cited limitations of PPG and metadata);
determining metric features from the meta data ; (See Park: Figs. 9 and 13, and [0148], “In particular embodiments, a cardiac rhythm inference system 910 may accept additional sources of data, generally described as alternate sensor channels, in addition to R-R interval time series data, to enhance the accuracy and/or value of the inferred results. One additional source of data includes user activity time series data, such as that measured by a 3-axis accelerometer concurrently with the R-R interval time series measurements. In addition, the system may accept other relevant metadata that may help to improve the accuracy of the rhythm analysis, such as user age, gender, indication for monitoring, pre-existing medical conditions, medication information, medical history and the like, and also information on the specific day and time range for each time series submitted to the system. Furthermore, the measurement device might also provide some measure of beat detection confidence, for example, for each R-Peak or for sequential time periods. This confidence measure would be based on analysis the recorded signal that, in typical embodiments, would not be recorded due to storage space and battery energy requirements. Finally, in the particular case that the R-R interval time series data are derived from an ECG signal, the system may accept additional signal features computed from the ECG. These features may include a time series of intra-beat interval measurements (such as the QT or PR interval, or QRS duration), or a time series of signal statistics such as the mean, median, standard deviation or sum of the ECG signal sample values within a given time period”. Note that the metadata related features of R-R, P-R, etc. are mapped to the metric features determined from the metadata)
determining a subject-specific metric statistic using the metric features from the meta data (See Rodriguez-Llorente: Fig. 1, and [0034], “In some embodiments, the processing equipment may generate value pairs each including a first sample point of the physiological signal and a second point of the physiological signal spaced apart by a particular spacing based on a calculated value. The processing equipment may determine a best fit linear relationship based on the plurality of value pairs, determine at least one statistical metric based on the linear relationship and the value pairs, and qualify or disqualify the calculated value based on the at least one statistical metric. In some embodiments, the statistical metric may include a standard error between the value pairs and the linear relationship, as well as a slope of the linear relationship. In some embodiments, qualifying or disqualifying the calculated value may include determining a value indicative of the probability, relative to a predetermined probability distribution function, of the at least one statistical metric being outside of a set of bounding values. In some embodiments, the processing equipment may determine at least one additional spacing other than the particular spacing”; Fig. 17 and [0305], “Step 1710 may include processing equipment determining an algorithm setting based on the area ratios of step 1708. In some embodiments, the processing equipment may analyze a sequence of ratio values to determine a metric. For example, the 25% largest value (e.g., larger than about 75% of ratio values and smaller than about 25% of ratio values) of a sequence of ratio values of positive areas to adjacent negative areas may be compared to a threshold. In a further example, the ratio corresponding to the largest 25% of a sequence of ratio values of positive areas to adjacent negative areas may be compared to a threshold value. In a further example, the average of the largest 25% of a sequence of ratio values of positive areas to adjacent negative areas may be compared to a threshold value. The ratio is expected to be near 1 for data not exhibiting dicrotic notches. For data exhibiting dicrotic notches, the ratios may exhibit two tiers due to the two-tiered shaped of troughs in the difference signal. A first tier will be somewhat close to 1, while the second tier will be significantly larger than 1. By picking the 25% value, the processing equipment will likely pick a value in the middle of second tier for dicrotic notches and therefore be high when dicrotic notches are present. However, when dicrotic notches are not present, the selected value is likely close to 1 because the areas ratios are all generally close to 1 (provided noise is sufficiently low). In some embodiments, the processing equipment may normalize the determined ratios, sort the normalized ratios into a sorted array, and select the value at one fourth of the length of the array of sorted ratios. In some embodiments, the processing equipment may compare the 25% value to the 75% value. For data exhibiting a dicrotic notch the 25% and 75% values should each lie in the middle of the two tiers, while for data not exhibiting a dicrotic notch, the values may be expected to be relatively similar”; and [0370], “In some embodiments, the processing equipment may apply any suitable statistical technique to the two sorted difference signals. For example, the processing equipment may apply a KS Test to the first and second portions by comparing the sorted difference signals to a predetermined distribution. In a further example, the processing equipment may use a function other than a line as a fitting reference. For example, the processing equipment may fit a polynomial of any order to the first and second difference signals, or any other suitable function, and compare the fitted functions to each other or to a reference function”. Note that the statistic metrics such as the linear relationship and value pairs, ratios of area, KS test to the predetermined distribution, etc. are mapped to determining the subject-specific metric statistic);
determining whether the metric features from the meta data are associated (See Park: Figs. 10-11, and [0161], “A wide variety of different types of ECG or comparable biological signal features may be extracted. For example, R-peak locations may be extracted. In certain embodiments, the R-peak locations are extracted via various methods such as: a Pan-Tompkins algorithm (Pan and Tompkins, 1985), providing a real-time QRS complex detection algorithm employing a series of digital filtering steps and adaptive thresholding, or an analog R-peak detection circuit comprising an R-peak detector consisting of a bandpass filter, a comparator circuit, and dynamic gain adjustment to locate R-peaks. The RR-intervals may be calculated from peak locations and used as the primary feature for rhythm discrimination. In embodiments, an R-peak overflow flag may be extracted. If more than a certain number of R-peaks were detected during a given time window such that not all data can be transmitted, a flag may be raised by the firmware. Such an extraction may be used to eliminate noisy segments from analysis, on the basis that extremely short intervals of R-R are not physiologically possible. With similar motivation, an R-peak underflow flag may be extracted to indicate an unrealistically long interval between successive R peaks, provided appropriate considerations for asystole are made in this evaluation. In an alternative implementation with the same goal, the lack of presence of R peaks in a prolonged interval could be associated with a confidence measure, which would describe the likelihood that the interval was clinical or artifact”; and Fig. 18, and [0185], “The Rules Database 14060 may, in some embodiments, store data (for example, instructions, preferences, profile) that establish parameters for the thresholds for analyzing the feature data. In some embodiments, one or more of the databases or data sources may be implemented using a relational database, such as Sybase, Oracle, CodeBase, MySQL, SQLite, and Microsoft® SQL Server, and other types of databases such as, for example, a flat file database, an entity-relationship database, and object-oriented database, NoSQL database, and/or a record-based database”) with a desired metric statistic (See Patwardhan: Figs. 3-4, and [0081], “In accordance with further aspects of the present technique, the state of the disease progression may be quantified. To that end, a distance metric between the normal statistics and the statistics measured from the test subject, such as the patient 102 is defined. For example, if the normal model is approximated by a Gaussian distribution around the mean values obtained from the normal measurements, then a suitable distance metric may be the Mahalanobis distance”; and [0050], “Additionally, the analysis module 308 is also configured to process the ultrasound images 302, the extracted articular bone surface and/or the segmented joint capsule region to determine an objective volumetric measure of inflammatory changes at the bone joint level. By way of example, the ultrasound images 302 may be processed to provide a direct and intuitive volumetric visualization of the inflamed tissues, volumetric quantification, and/or volumetric Doppler flow assessment. Moreover, the analysis module 308 may also be configured to generate a score or a metric corresponding to the disease state, where the score or metric provides a direct joint-level assessment of the disease state at a point of care, such as a rheumatologist's office. In one embodiment, the analysis module 308 may be configured to determine the score or metric, the disease state and the like by comparing the determined score or metric with measurements corresponding to normal subjects. The normal measurements may be stored in a normal model database 314, for example”. Note that the normal measurements are mapped to the meta data metric features, and the normal statistics of the normal measurements is mapped to the desired metric statistics) such that the subject-specific metric statistic is expected to meet the desired metric statistic (See Ochs: Fig. 1, and [0015], “Physiological monitoring systems rely on signals that are often corrupted or contain morphologies that produce erratic or unreliable results when processing occurs to extract useful physiological information. The present disclosure describes a system that addresses these concerns by identifying portions of a physiological signal, such as a photoplethysmograph (PPG) signal, that, when processed, are likely to result in erratic or unreliable results”; [0018], “The patient monitoring system may identify the largest contiguous portion of the PPG signal that does not correspond to a large baseline shift and further identify an additional buffer region. The remaining portions of the PPG signal (including the buffer region) may be discarded, replaced, or otherwise ignored. The patient monitoring system may also identify the largest contiguous portion of the PPG signal that does not correspond to an artifact, large pulse-to-pulse variability, or out of range values. The contiguous portion may be required to be of a particular minimum length. The remaining portions of the PPG signal may be discarded, replaced, or otherwise ignored. The remaining usable portion of the PPG signal may be used to determine physiological information such as respiration information”; and Fig. 5, and [0065], “FIG. 5 depicts a flow diagram showing illustrative steps for determining a usable portion of a physiological signal such as a PPG signal in accordance with some embodiments of the present disclosure. Although a number of exemplary steps are disclosed herein, it will be understood that any of the steps depicted in FIG. 5 may be omitted, the order of the steps may be modified, and additional steps may be added. For example, any of the steps for determining signal portions (i.e., steps 502, 504, 506, 508, and 510) may be omitted, and additional signal portion identifying steps may be added. The steps including identifying contiguous signal portions may be performed as depicted in FIG. 5 (i.e., steps 502 and 512), after all signal portions are identified (e.g., at step 512 only), or may be performed individually after each signal portion is identified. Similarly, modifying the determined signal portions may be performed as depicted in FIG. 5 (i.e., at step 512), after some of the signal portions are identified (e.g., at steps 502 and 512), or may be performed individually after each signal portion is identified. In some embodiments different signal modification procedures may be associated with different signal portion determination steps”. Note that extracting features from the PPG signals, using these features to identify the usable portion of the PPG signa, and using these usable portion of PPG signal to determine the physiological information, is mapped to the current claimed limitation of “a desired metric statistic such that the subject-specific metric statistic is expected to meet the desired metric statistic”); and
in response to determining that the metric features from the meta data are associated with (See Park: Fig. 11, and [0167], “The identified arrhythmia locations 1018 are then transmitted 1020 back to the sensor 1002. The transmission 1020 back to the sensor may be accomplished by any communication protocols/technology described herein this section or elsewhere in the specification, for example via Bluetooth. The sensor then reads the transmitted identified locations 1022 and accesses 1024 the areas of memory corresponding to the transmitted identified locations 1022 of the ECG”. Note that the identified location for some events or measurements such as R-R are mapped to the metric features associated with a preset parameter) the desired metric statistic (See Patwardhan: Figs. 3-4, and [0081], “In accordance with further aspects of the present technique, the state of the disease progression may be quantified. To that end, a distance metric between the normal statistics and the statistics measured from the test subject, such as the patient 102 is defined. For example, if the normal model is approximated by a Gaussian distribution around the mean values obtained from the normal measurements, then a suitable distance metric may be the Mahalanobis distance”; and [0050], “Additionally, the analysis module 308 is also configured to process the ultrasound images 302, the extracted articular bone surface and/or the segmented joint capsule region to determine an objective volumetric measure of inflammatory changes at the bone joint level. By way of example, the ultrasound images 302 may be processed to provide a direct and intuitive volumetric visualization of the inflamed tissues, volumetric quantification, and/or volumetric Doppler flow assessment. Moreover, the analysis module 308 may also be configured to generate a score or a metric corresponding to the disease state, where the score or metric provides a direct joint-level assessment of the disease state at a point of care, such as a rheumatologist's office. In one embodiment, the analysis module 308 may be configured to determine the score or metric, the disease state and the like by comparing the determined score or metric with measurements corresponding to normal subjects. The normal measurements may be stored in a normal model database 314, for example”. Note that the normal measurements are mapped to the meta data metric features, and the normal statistics of the normal measurements is mapped to the desired metric statistics),
categorizing the subject-specific metric statistic as qualified for processing a PPG-based metric for the subject (See Park: Fig. 11, and [0167], “The identified arrhythmia locations 1018 are then transmitted 1020 back to the sensor 1002. The transmission 1020 back to the sensor may be accomplished by any communication protocols/technology described herein this section or elsewhere in the specification, for example via Bluetooth. The sensor then reads the transmitted identified locations 1022 and accesses 1024 the areas of memory corresponding to the transmitted identified locations 1022 of the ECG. In some embodiments, the sensor applies additional analysis of the identified segments to further build confidence in the arrhythmia identification. This further rhythm confidence determination step 1026 allows for increasing positive predictivity prior to the power-hungry transmission step. In embodiments, if the confidence exceeds a defined threshold the data segment is transmitted. For example, the defined threshold may be a preset value or it may be set per user and monitoring session. In embodiments, the defined threshold may be changed dynamically depending on the nature of the rhythm, the history of accurate detection within the monitoring period, and/or the confidence of the rhythm inference system. Additional analysis may also be performed. Examples of possible analysis techniques include any methods disclosed herein this section or elsewhere in the specification, for example: R-peak amplitude, ECG signal amplitude proxy, ECG signal samples, local ECG signal energy, spectral information, and/or output from a simple machine-learned model”; and [0168], “If the confidence exceeds a threshold as described above, the sensor 1002 may transmit the requested ECG segments 1028 to the processing device via any transmission means described herein this section or elsewhere in the specification. The processing device may complete further analysis on the segments to confirm accuracy of predicted arrhythmia before using data to report to a user and/or physician, as needed”. Note that the arrhythmia is mapped to the category of metric feature, the confidence arrhythmia is mapped to the user-specific metric feature, and threshold exceeded is mapped to it is qualified for further process, and transmitting to the processing unit for additional analysis is mapped to processing the PPG).
Claims 2-3 are rejected under 35 U.S.C. 103 as being unpatentable over Park, etc. (US 20160120434 A1) in view of Patwardhan, etc. (US 20130060121 A1), further in view of Rodriguez-Llorente, etc. (US 20140073865 A1), Ochs, etc. (US 20130289413 A1), and Garcia Molina, etc. (US 20100090798 A1).
Regarding claim 2, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach all the features with respect to claim 1 as outlined above. However, Park, modified by Patwardhan, Rodriguez-Llorente, and Ochs, fails to explicitly disclose that the method of claim 1, wherein the meta data comprises subject-specific biometric information.
However, Garcia Molina teaches that the method of claim 1, wherein the meta data comprises subject-specific biometric information (See Garcia Molina: Fig. 2, and [0014], “A basic idea of the invention is that, rather than determining peak locations in cyclic signals such as ECG or PPG signals when using these signals as a representation of biometric data for verifying identity of an individual, shape or morphology of the signals is considered. In PQRST cycles forming an ECG, the morphology of R-R segments can be used as a means for comparison between a biometric measurement and a biometric template. Whereas the relative location of distinctive patterns in a PQRST cycle can change, the morphology of R-R segments remains essentially unchanged. Typically, the R-peaks are taken as reference because they are present in every electrode configuration and can be more precisely and unambiguously determined as they constitute the highest peaks in the ECG signal. Also, all the elements of a PQRST-cycle are contained within an R-R segment. Even though identification of an individual by means of extracting feature data sets from the R-R segment is discussed throughout this description, it should be clearly understood by a skilled person that other segments could be considered, as well as other suitable signals from which the segments are selected. Further, to improve performance of the biometric identification, a sequence of R-R segments may be employed in the verification procedure”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention was effectively filed to modify Park to have the method of claim 1, wherein the meta data comprises subject-specific biometric information as taught by Garcia Molina in order to efficiently verify the identity of the individual by employing the biometric data derived from the physical features of the individual (See Garcia Molina: Fig. 1, and [0010], “This object is attained by a method of verifying identity of an individual by employing biometric data derived from a physical feature of the individual in accordance with claim 1 and a device for verifying identity of an individual by employing biometric data derived from a physical feature of the individual in accordance with claim 9”). Park teaches a method and system that may monitor the subjects by detecting multiple subject physiological states/measurements such as PPG, ECG, etc., using various sensors, extracting features from the PPG and ECG signals, analyzing the features with comparing the features to the thresholds, and feedbacking the analysis results to the users; while Garcia Molina teaches a system and method that may identify the user identification using the biometric data to distinguish the signal group. Therefore, it is obvious to one of ordinary skill in the art to modify Park by Garcia Molina to identify user identification using the biometric data. The motivation to modify Park by Garcia Molina is “Use of known technique to improve similar devices (methods, or products) in the same way”.
Regarding claim 3, Park, Patwardhan, Rodriguez-Llorente, Ochs, and Garcia Molina teach all the features with respect to claim 2 as outlined above. Further, Park teaches that the method of claim 2, wherein the subject-specific biometric information comprises one or more of subject height, subject weight, subject body mass index, subject age, subject gender, subject ethnicity, subject skin tone, or subject medication usage (See Park: Fig. 1, and [0148], “In particular embodiments, a cardiac rhythm inference system 910 may accept additional sources of data, generally described as alternate sensor channels, in addition to R-R interval time series data, to enhance the accuracy and/or value of the inferred results. One additional source of data includes user activity time series data, such as that measured by a 3-axis accelerometer concurrently with the R-R interval time series measurements. In addition, the system may accept other relevant metadata that may help to improve the accuracy of the rhythm analysis, such as user age, gender, indication for monitoring, pre-existing medical conditions, medication information, medical history and the like, and also information on the specific day and time range for each time series submitted to the system. Furthermore, the measurement device might also provide some measure of beat detection confidence, for example, for each R-Peak or for sequential time periods. This confidence measure would be based on analysis the recorded signal that, in typical embodiments, would not be recorded due to storage space and battery energy requirements. Finally, in the particular case that the R-R interval time series data are derived from an ECG signal, the system may accept additional signal features computed from the ECG. These features may include a time series of intra-beat interval measurements (such as the QT or PR interval, or QRS duration), or a time series of signal statistics such as the mean, median, standard deviation or sum of the ECG signal sample values within a given time period”).
Claims 5-8 are rejected under 35 U.S.C. 103 as being unpatentable over Park, etc. (US 20160120434 A1) in view of Patwardhan, etc. (US 20130060121 A1), further in view of Rodriguez-Llorente, etc. (US 20140073865 A1), Ochs, etc. (US 20130289413 A1), and Garcia Addison, etc. (US 20170105672 A1).
Regarding claim 5, Park, Patwardhan, Rodriguez-Llorente, and Ochs teach all the features with respect to claim 1 as outlined above. However, Park, modified by Patwardhan, Rodriguez-Llorente, and Ochs fails to explicitly disclose that the method of claim 1, wherein determining whether the metric features from the meta data are associated with the desired metric statistic comprises determining whether the metric features from the meta data belong to a feature cluster associated with the desired metric statistic.
However, Addison teaches that the method of claim 1, wherein determining whether the metric features from the meta data are associated with the desired metric statistic comprises determining whether the metric features from the meta data belong to a feature cluster associated with the desired metric statistic (See Addison: Figs. 1-5, and [0018], “FIG. 1 is a block diagram of an embodiment of a system 10 for monitoring a patient's autoregulation. As shown, the system 10 includes a blood pressure sensor 12, an oxygen saturation sensor 14 (e.g., a regional oxygen saturation sensor), a controller 16, and an output device 18. The blood pressure sensor 12 may be any sensor or device configured to obtain the patient's blood pressure (e.g., mean arterial blood pressure (MAP)). For example, the blood pressure sensor 12 may include a blood pressure cuff for non-invasively monitoring blood pressure or an arterial line for invasively monitoring blood pressure. In certain embodiments, the blood pressure sensor 12 may include one or more pulse oximetry sensors. In some such cases, the patient's blood pressure may be derived by processing time delays between two or more characteristic points within a single plethysmography (PPG) signal obtained from a single pulse oximetry sensor. Various techniques for deriving blood pressure based on a comparison of time delays between certain components of a single PPG signal obtained from a single pulse oximetry sensor is described in U.S. Publication No. 2009/0326386, entitled “Systems and Methods for Non-Invasive Blood Pressure Monitoring,” the entirety of which is incorporated herein by reference. In other cases, the patient's blood pressure may be continuously, non-invasively monitored via multiple pulse oximetry sensors placed at multiple locations on the patient's body. As described in U.S. Pat. No. 6,599,251, entitled “Continuous Non-invasive Blood Pressure Monitoring Method and Apparatus,” the entirety of which is incorporated herein by reference, multiple PPG signals may be obtained from the multiple pulse oximetry sensors, and the PPG signals may be compared against one another to estimate the patient's blood pressure. Regardless of its form, the blood pressure sensor 12 may be configured to generate a blood pressure signal indicative of the patient's blood pressure (e.g., arterial blood pressure) over time. As discussed in more detail below, the blood pressure sensor 12 may provide the blood pressure signal to the controller 16 or to any other suitable processing device to enable identification of the autoregulation zone(s) and to enable evaluation of the patient's autoregulation status”; and [0024], “FIG. 3 is an example of a graph 50 illustrating the COx 54 plotted against blood pressure 56 (e.g., mean arterial pressure (MAP)). In particular, the graph 50 of FIG. 3 illustrates individual raw data points 58. As shown, the data points 58 are distributed (e.g., spread) across COx values 54 in a characteristic manner at the various blood pressures 56. In particular, the data points 58 may have a relatively greater spread across COx values 54 at intermediate blood pressures associated with an intact autoregulation zone 60. Additionally, the data points 58 may have a relatively lower spread across COx values 54 at lower blood pressures associated with a lower impaired autoregulation zone 62 and at higher blood pressures associated with a higher impaired autoregulation zone 64. Furthermore, the data points 58 may generally vary between −1 and +1 at the intermediate blood pressures associated with the intact autoregulation zone 60, and may cluster at approximately +1 at the lower blood pressures associated with the lower impaired autoregulation zone 62 and at the higher blood pressures associated with the higher impaired autoregulation zone 64. These distribution patterns and/or characteristics may be utilized to facilitate efficient and/or reliable determination of the various autoregulation zones, the LLA, the ULA, and/or a target blood pressure. For example, any of a variety of data clustering algorithms may be utilized by the controller 16 to cluster the data points 58, thereby facilitating identification of the autoregulation zones, the LLA, the ULA, and/or a target blood pressure, as discussed in detail below”. Note that group is mapped to cluster).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filling date of the claimed invention was effectively filed to modify Park to have the method of claim 1, wherein determining whether the metric features from the meta data are associated with the desired metric statistic comprises determining whether the metric features from the meta data belong to a feature cluster associated with the desired metric statistic as taught by Addison in order to enable reliable determination of patient's auto-regulation status in an efficient manner (See Addison: Fig. 1, and [0017], “Accordingly, systems and methods for efficiently and/or reliably identifying the ULA, LLA, and/or blood pressures associated with the autoregulation zones, and thereby enabling efficient and/or reliable determination of the patient's autoregulation status, are provided herein. Furthermore, in some embodiments, the systems and methods may be configured to determine a target blood pressure for the patient. In some embodiments, the target blood pressure may be a blood pressure value or a range of blood pressure values within the intact autoregulation zone”). Park teaches a method and system that may monitor the subjects by detecting multiple subject physiological states/measurements such as PPG, ECG, etc., using various sensors, extracting features from the PPG and ECG signals, analyzing the features with comparing the features to the thresholds, and feedbacking the analysis results to the users; while Addison teaches a system and method that may extract the physiological features from the PPG or other medical signal by clustering the signals and analyzing the signal clusters. Therefore, it is obvious to one of ordinary skill in the art to modify Park by Addison to process the PPG signals by clustering the signals and analyzing the signals clusters. The motivation to modify Park by Addison is “Use of known technique to improve similar devices (methods, or products) in the same way”.
Regarding claim 6, Park, Patwardhan, Rodriguez-Llorente, Ochs, and Addison teach all the features with respect to claim 5 as outlined above. Further, Addison teaches that the method of claim 5, wherein the feature cluster is one of a plurality of metric feature clusters generated using one or more data clustering techniques (See Addison: Fig. 5, and [0029], “FIG. 5 is an example of a graph 80 illustrating application of a k-means clustering algorithm to the data points 58 of the graph 70 of FIG. 4. To generate the graph 80, the k-means clustering algorithm is applied (e.g., by the controller 16) to the raw data points 58 without or prior to data binning (e.g., without or prior to grouping the raw data points 58 into a smaller number of bins or blood pressure intervals). As shown, application of the k-means clustering algorithm to the data points 58 may result in a first cluster 82 associated with lower blood pressures and a second cluster 84 associated with higher blood pressures. In the illustrated graph 80, each of the data points 58 within the first cluster 82 are represented by circles, and each of the data points 58 within the second cluster 84 are represented by crosses. The controller 16 may evaluate the clusters 82, 84 (e.g., evaluate blood pressures within the clusters 82, 84, etc.) to determine that the first cluster 82 corresponds to the lower impaired autoregulation zone 62, and the second cluster 84 corresponds to the intact autoregulation zone 60, in the illustrated embodiment. Furthermore, the controller 16 may determine a boundary 86 between the clusters 82, 84, and may determine that the boundary 86 corresponds to the LLA”).
Regarding claim 7, Park, Patwardhan, Rodriguez-Llorente, Ochs, and Addison teach all the features with respect to claim 6 as outlined above. Further, Addison teaches that the method of claim 6, wherein the one or more data clustering techniques comprise k-means clustering or a Gaussian mixture model (See Addison: Fig. 6, Nd [0037], “FIG. 6 is an example of a graph 90 after application of a Gaussian mixture model to the data points 58 of the graph 70 of FIG. 4. To generate the graph 90, the Gaussian mixture model is applied (e.g., by the controller 16) to the raw data points 58 without or prior to data binning. As shown, application of the Gaussian mixture model to the data points 58 may result in the first cluster 82 associated with lower blood pressures and the second cluster 84 associated with higher blood pressures. In the illustrated graph 90, the first cluster 82 is marked by a first set of ellipsoids 92, and the second cluster 84 is marked by a second set of ellipsoids 94. As discussed above, the controller 16 may evaluate the clusters 82, 84 (e.g., evaluate blood pressures within the clusters 82, 84, etc.). For example, in the illustrated embodiment, the controller 16 may evaluate the clusters 82, 84 to determine that the first cluster 82 corresponds to the lower impaired autoregulation zone 62, and the second cluster 84 corresponds to the intact autoregulation zone 60. Furthermore, the controller 16 may determine a boundary 86 between the clusters 82, 84, and may determine that the boundary 86 corresponds to the LLA. In the illustrated embodiment, the boundary 86 is located at a midpoint between the clusters 82, 84 (e.g., between adjacent edges of the clusters 82, 84). In some embodiments, the boundary 86 may be located at an edge of the first cluster 82 associated with the intact autoregulation zone 60 or between the midpoint and the edge of the first cluster 82, for example”).
Regarding claim 8, Park, Patwardhan, Rodriguez-Llorente, Ochs, and Addison teach all the features with respect to claim 5 as outlined above. Further, Park teaches that the method of claim 5, wherein the feature cluster is associated with an accuracy value for the PPG-based metric (See Park: Fig. 9, and [0144], “The R-R interval time series 902 data may be extracted from or received from a dedicated heart rate monitor such as a heart rate chest strap or heart rate watch, or a wearable health or fitness device 906, 908 that incorporates heart rate sensing functionality. Alternatively, the R-R interval time series 902 may be derived from a wearable patch designed to measure an ECG signal 904 (for instance, by locating the R peaks in the ECG using a QRS detection algorithm). Furthermore, the R-R interval time series 902 may be estimated from an alternative physiological signal such as that obtained from photoplethysmography (PPG). In this scenario, the peak-to-peak interval time series determined from the PPG signal may be used as an accurate estimate of the R-R interval time series”).
Allowable Subject Matter
Claim 13 is objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims. The best arts searched, Park, etc. (US 20160120434 A1), Patwardhan, etc. (US 20130060121 A1), Rodriguez-Llorente, etc. (US 20140073865 A1), Ochs, etc. (US 20130289413 A1), and Garcia Molina, etc. (US 20100090798 A1), and Addison, etc. (US 20170105672 A1), do not teach the cited limitation of “the method of claim 12, wherein the error for the blood pressure metric comprises an error of< 5 ± 8mmHG for the blood pressure metric.”
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 GORDON G LIU whose telephone number is (571)270-0382. The examiner can normally be reached Monday - Friday 8:00-5:00.
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, Devona E Faulk can be reached at 571-272-7515. 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.
/GORDON G LIU/Primary Examiner, Art Unit 2618
Prosecution Insights