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 .
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
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.
Claim(s) 1, 15, 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chung et al. (KR 20220077495 A) in views Zou (CN 118041543 A).
In regards to claim 1, Chung teaches a computer implemented method of determining an identity of an authorized user, the computer implemented method comprising: receiving a first identity token from an authorized user, obtaining, at the time of or after receiving the first identity token, a first electrocardiogram (ECG) signal for the authorized user(Abstract; Page 3, Paragraph 4)
Disclosed are an identity verification apparatus for remotely measured biometric data and a method using the same. In an identity verification method according to an embodiment of the present invention, an identity verification device obtains a target ECG signal remotely measured for an authenticated user based on a wearable medical device, and based on a machine learning-based identity inference engine extracts at least one candidate user estimated to be a user of the target ECG signal from among a plurality of users registered in the authentication information database, and a reference feature vector of the at least one candidate user and a plurality of individual extracted from the target ECG signal The waveforms are compared, and the identity verification of the authenticated user is performed based on the identification result of each of the plurality of individual waveforms determined according to the comparison result.[Abstr]
Therefore, the present invention was focused on the point that a patient's ECG signal has the characteristic of identifying an individual, and using this, a technical link that combines the ECG signal and the patient's identity can be made. That is, as shown in FIG. 2 , the identity of the source patient is identified using physiological data that changes with time and the measurement value is not constant at each measurement, and that the patient is the person corresponding to the monitoring target.[Pg 3. P-4]
For example, a new user visits a hospital face-to-face or non-face-to-face, measures his/her ECG signal based on his/her own or distributed Telehealth/Telemedicine service device and smartphone application, and registers it in the authentication information database, An identifier (ID) may be assigned corresponding to the patient number.[Pg 5, P-8]
Here, Chung discloses the computer implemented method receiving a first identity token from an authorized user, i.e. measures his/her ECG signal based on his/her own or distributed Telehealth/Telemedicine service device and smartphone application, and registers it in the authentication information database, An identifier (ID) may be assigned corresponding to the patient number. Furthermore, an identity verification device obtains a target ECG signal remotely measured for an authenticated user based on a wearable medical device, and based on a machine learning-based identity inference engine extracts at least one candidate user estimated to be a user of the target ECG signal from among a plurality of users registered in the authentication information database.
Furthermore, Chung teaches providing at least part of the first ECG signal to a machine learning algorithm to obtain a first vector representation of first features of the first ECG signal, labelling the first vector representation as belonging to the authorized user, the labelling being based upon the first identity token(Abstract; Page 3, Paragraph 4)
Disclosed are an identity verification apparatus for remotely measured biometric data and a method using the same. In an identity verification method according to an embodiment of the present invention, an identity verification device obtains a target ECG signal remotely measured for an authenticated user based on a wearable medical device, and based on a machine learning-based identity inference engine extracts at least one candidate user estimated to be a user of the target ECG signal from among a plurality of users registered in the authentication information database, and a reference feature vector of the at least one candidate user and a plurality of individual extracted from the target ECG signal The waveforms are compared, and the identity verification of the authenticated user is performed based on the identification result of each of the plurality of individual waveforms determined according to the comparison result.[Abstr]
Therefore, the present invention was focused on the point that a patient's ECG signal has the characteristic of identifying an individual, and using this, a technical link that combines the ECG signal and the patient's identity can be made. That is, as shown in FIG. 2 , the identity of the source patient is identified using physiological data that changes with time and the measurement value is not constant at each measurement, and that the patient is the person corresponding to the monitoring target.[Pg 3. P-4]
For example, a new user visits a hospital face-to-face or non-face-to-face, measures his/her ECG signal based on his/her own or distributed Telehealth/Telemedicine service device and smartphone application, and registers it in the authentication information database, An identifier (ID) may be assigned corresponding to the patient number.[Pg 5, P-8]
First, when a target electrocardiogram signal (ECG) for a user authenticated based on a wearable medical device is received (S1010), a target feature vector (W') is extracted from the target electrocardiogram signal (S1020), and a target feature vector (W') ) can be input to the identity inference engine to derive target public information (S) (S1030).[Pg 9, P-3]
Thereafter, Chung teaches obtaining a second ECG signal for a first candidate user, providing at least part of the second ECG signal to the machine learning algorithm to obtain a second vector representation of second features of the second ECG signal; comparing the second vector representation to the labelled first vector representation to determine a similarity metric; and if the similarity metric is within a predetermined threshold, providing confirmation that the first candidate driver is the authorized driver. (Abstract; Page 10, Paragraphs 5-11; Page 19, Paragraph 9-Page 20, Paragraph 7)
Disclosed are an identity verification apparatus for remotely measured biometric data and a method using the same. In an identity verification method according to an embodiment of the present invention, an identity verification device obtains a target ECG signal remotely measured for an authenticated user based on a wearable medical device, and based on a machine learning-based identity inference engine extracts at least one candidate user estimated to be a user of the target ECG signal from among a plurality of users registered in the authentication information database, and a reference feature vector of the at least one candidate user and a plurality of individual extracted from the target ECG signal The waveforms are compared, and the identity verification of the authenticated user is performed based on the identification result of each of the plurality of individual waveforms determined according to the comparison result. [Abstr]
Therefore, as a result of the determination in step S1055, when the error correction accuracy for the corrected target ECG signal is verified by matching the target verification hash value EV .sub.i '' and the verification hash value EV .sub.i , the correction target characteristic The vector W'' may be input to the identity inference engine to generate a candidate identity list (EID LIST) including at least one candidate user ( S1060 ). At this time, the corrected target feature vector (W'') by the successfully calibrated target ECG signal is input to the identity inference engine, and ID .sub.i <- Estimate(W'') based on Equation 6 is used to The user's identity can be inferred. For example, the identity inference engine identifies at least one candidate user (EID) having a predicted probability value as in [Equation 9], and generates a candidate identity list corresponding to the identified at least one candidate user (EID). Can [Pg 10, P-5-11]
In addition, the processor 1220 compares the reference feature vector of at least one candidate user and a plurality of individual waveforms extracted from the target ECG signal. In addition, the processor 1220 performs identity verification on the authenticated user based on the identification result for each of the plurality of individual waveforms determined according to the comparison result. That is, it is possible to verify whether the target ECG signal remotely measured for an authenticated user based on the wearable medical device comes from the patient's own body, which is expected to be monitored, and whether the person's identity is the same. Hereinafter, a process of verifying the source identity of the target ECG signal based on at least one candidate user (EID) estimated through Equation 9 will be described in detail with reference to FIG. 11. First, when a target electrocardiogram signal (ECG) remotely measured for a user (Real ID) authenticated based on a wearable medical device is received (S1110), at least one candidate as shown in Equation 9 using an identity inference engine The user (EID = {eID .sub.11 , ..., eID .sub.kn } may be extracted (S1120). In this case, the authenticated user's identifier (Real ID) may be obtained in a user authentication process for establishing a service session between a user using the wearable medical device and the identity verification device according to the present invention. In this case, the type of the identifier (Real ID) may not be limited. For example, a patient number that can be linked to a hospital's medical information system may be used as an identifier (Real ID). If a patient number that can be identified as a barcode or QR code can be extracted from the patient's ECG signal, an expected effect that can be utilized by easily linking with various hospital medical information systems linked to the patient number can occur. Thereafter, identity verification may be performed by comparing at least one candidate user (EID = {eID .sub.11 , ..., eID .sub.kn } with the authenticated user (Real ID) (S1130). In this case, a comparison check may be performed on both the authenticated user (Real ID) and at least one candidate user (eID .sub.ij ) identified in [Equation 9] until identity verification is successful. Thereafter, it is possible to perform a comparative test on each of the plurality of individual waveforms extracted from the target ECG signal, and determine whether identity verification is successful for all of the plurality of individual waveforms extracted from the target ECG signal ( S1135). In this case, it may be determined that the identification of the individual waveform matching the reference feature vector among the plurality of individual waveforms is successful.[Pg 19, P-9-Pg 20, P-7]
Here we see, Chung’s teaching after acquiring the token ID of a user affiliated with their ECG vector. The reference vector is stored in a database as the users ID. Thereafter, another candidate vector ID is acquired and compared accordingly to match the reference vector to determine if the user is an authorized user’s ID with respect to the reference vector.
Chung, however, fails to teach the authorization process being associated with, authorizing and verifying an authorized driver’s identity.
Zou on the other hand teaches obtaining a drivers biometric ID converted and stored in a database/server to which when a candidate vector ins read and compared to the stored authorized driver’s biometric vector ID, a match is determined, upon which, access to operate the vehicle is granted (Abstract; Page 2, Paragraph 5; Page 3, Paragraph 4)
The invention claims a vehicle-mounted system and operation method thereof. The vehicle-mounted system comprises a data obtaining device, a biological characteristic obtaining device and a processor. The data acquisition device is used for acquiring a self-key generated by de-identifying a first biological feature of a user using the vehicle to obtain first de-identified data and converting the first de-identified data into a first feature vector including a plurality of first de-identified features. The biological feature obtaining device is used for obtaining the second biological feature of the current user to be identified. The processor is used for de-identifying the second biological characteristic to obtain the second de-identified data, and converting it into the second characteristic vector containing multiple second de-identified characteristics, and comparing it with the first characteristic vector in the self key to start the pre-set function of the vehicle according to the comparison result.[Abstr]
The invention claims a vehicle system configured in a vehicle, comprising a data obtaining device, a first biological characteristic obtaining device and a processor. wherein the data obtaining device is used for obtaining the registered self-key, wherein the self-key is used for obtaining the first de-identification data by performing de-identification processing on the first biological characteristic of the user using the vehicle, and converting the first de-identified data into a first eigenvector containing a plurality of first de-identified features. The first biological characteristic obtaining device is used for obtaining the second biological characteristic of the current user to be identified. the processor is coupled with the data obtaining device and the first biological feature obtaining device, configured to perform de-identification processing on the second biological feature to obtain the second de-identification data, and convert the second de-identification data into a second feature vector comprising a plurality of second de-identification features, and comparing it with the first characteristic vector in the self key to start the predetermined function of the vehicle according to the comparison result.[Pg 2, P-5]
In some embodiments, the first biometric acquisition device is configured outside a vehicle for acquiring a third biometric of an external user. wherein the processor further performs de-identification processing on the third biological feature to obtain the third de-identification data, and converts the third de-identification data into the third feature vector containing a plurality of third de-identification features, and compares it with the first feature vector in its own key, The door of the vehicle is opened according to the comparison result.[Pg 3, P-4]
It would therefore be obvious during the time of the filing date to combine Zou’s teaching with Chung’s teaching in order to enable a more efficient and effective method to determine an authorized driver/operator of a vehicle, accordingly.
In regards to claim 15, Chung teaches a system for determining an identity of a user, the system comprising an identity token comprising information identifying the user; at least one electrocardiogram (ECG) electrode configured to be contacted by a body part of the user and to output an ECG signal(Abstract; Page 3, Paragraph 4)
Disclosed are an identity verification apparatus for remotely measured biometric data and a method using the same. In an identity verification method according to an embodiment of the present invention, an identity verification device obtains a target ECG signal remotely measured for an authenticated user based on a wearable medical device, and based on a machine learning-based identity inference engine extracts at least one candidate user estimated to be a user of the target ECG signal from among a plurality of users registered in the authentication information database, and a reference feature vector of the at least one candidate user and a plurality of individual extracted from the target ECG signal The waveforms are compared, and the identity verification of the authenticated user is performed based on the identification result of each of the plurality of individual waveforms determined according to the comparison result.[Abstr]
Therefore, the present invention was focused on the point that a patient's ECG signal has the characteristic of identifying an individual, and using this, a technical link that combines the ECG signal and the patient's identity can be made. That is, as shown in FIG. 2 , the identity of the source patient is identified using physiological data that changes with time and the measurement value is not constant at each measurement, and that the patient is the person corresponding to the monitoring target.[Pg 3. P-4]
For example, a new user visits a hospital face-to-face or non-face-to-face, measures his/her ECG signal based on his/her own or distributed Telehealth/Telemedicine service device and smartphone application, and registers it in the authentication information database, An identifier (ID) may be assigned corresponding to the patient number.[Pg 5, P-8]
Here Chung teaches an identity token comprising information identifying the user at least one electrocardiogram (ECG) electrode configured to be contacted by a body part of a user outputting an ECG signal, i.e. measures his/her ECG signal based on his/her own or distributed Telehealth/Telemedicine service device and smartphone application, and registers it in the authentication information database.
Furthermore, Chung teaches an ECG processor comprising an embedding module, the embedding module comprising a machine learning algorithm, the embedding module being configured to receive at least part of the ECG signal and to use the machine learning algorithm to obtain a vector representation of features of the ECG signal (Abstract; Page 3, Paragraph 4)
Disclosed are an identity verification apparatus for remotely measured biometric data and a method using the same. In an identity verification method according to an embodiment of the present invention, an identity verification device obtains a target ECG signal remotely measured for an authenticated user based on a wearable medical device, and based on a machine learning-based identity inference engine extracts at least one candidate user estimated to be a user of the target ECG signal from among a plurality of users registered in the authentication information database, and a reference feature vector of the at least one candidate user and a plurality of individual extracted from the target ECG signal The waveforms are compared, and the identity verification of the authenticated user is performed based on the identification result of each of the plurality of individual waveforms determined according to the comparison result.[Abstr]
Therefore, the present invention was focused on the point that a patient's ECG signal has the characteristic of identifying an individual, and using this, a technical link that combines the ECG signal and the patient's identity can be made. That is, as shown in FIG. 2 , the identity of the source patient is identified using physiological data that changes with time and the measurement value is not constant at each measurement, and that the patient is the person corresponding to the monitoring target.[Pg 3. P-4]
For example, a new user visits a hospital face-to-face or non-face-to-face, measures his/her ECG signal based on his/her own or distributed Telehealth/Telemedicine service device and smartphone application, and registers it in the authentication information database, An identifier (ID) may be assigned corresponding to the patient number.[Pg 5, P-8]
First, when a target electrocardiogram signal (ECG) for a user authenticated based on a wearable medical device is received (S1010), a target feature vector (W') is extracted from the target electrocardiogram signal (S1020), and a target feature vector (W') ) can be input to the identity inference engine to derive target public information (S) (S1030).[Pg 9, P-3]
Here, we see an ECG processor comprising an embedding module, the embedding module comprising a machine learning algorithm, the embedding module being configured to receive at least part of the ECG signal and to use the machine learning algorithm to obtain a vector representation of features of the ECG signal, i.e. measures his/her ECG signal based on his/her own or distributed Telehealth/Telemedicine service device and smartphone application, and registers it in the authentication information database, An identifier (ID) may be assigned corresponding to the patient number; when target electrocardiogram signal (ECG) for a user authenticated based on a wearable medical device is received (S1010), a target feature vector (W') is extracted from the target electrocardiogram signal (S1020), and a target feature vector (W') ) can be input to the identity inference engine to derive target public information (S) (S1030).
Thereafter, a classifier configured to compare two vector representations to determine a similarity metric; wherein the system is configured to: receive a first identity token from an authorized user; obtain, at the time of or after receiving the first identity token, a first ECG signal for the authorized user; provide at least part of the first ECG signal to the machine learning algorithm to obtain a first vector representation of first features of the first ECG signal; label the first vector representation as belonging to the authorized driver, the labelling being based upon the first identity token(Abstract; Page 3, Paragraph 4)
Disclosed are an identity verification apparatus for remotely measured biometric data and a method using the same. In an identity verification method according to an embodiment of the present invention, an identity verification device obtains a target ECG signal remotely measured for an authenticated user based on a wearable medical device, and based on a machine learning-based identity inference engine extracts at least one candidate user estimated to be a user of the target ECG signal from among a plurality of users registered in the authentication information database, and a reference feature vector of the at least one candidate user and a plurality of individual extracted from the target ECG signal The waveforms are compared, and the identity verification of the authenticated user is performed based on the identification result of each of the plurality of individual waveforms determined according to the comparison result.[Abstr]
Therefore, the present invention was focused on the point that a patient's ECG signal has the characteristic of identifying an individual, and using this, a technical link that combines the ECG signal and the patient's identity can be made. That is, as shown in FIG. 2 , the identity of the source patient is identified using physiological data that changes with time and the measurement value is not constant at each measurement, and that the patient is the person corresponding to the monitoring target.[Pg 3. P-4]
For example, a new user visits a hospital face-to-face or non-face-to-face, measures his/her ECG signal based on his/her own or distributed Telehealth/Telemedicine service device and smartphone application, and registers it in the authentication information database, An identifier (ID) may be assigned corresponding to the patient number.[Pg 5, P-8]
First, when a target electrocardiogram signal (ECG) for a user authenticated based on a wearable medical device is received (S1010), a target feature vector (W') is extracted from the target electrocardiogram signal (S1020), and a target feature vector (W') ) can be input to the identity inference engine to derive target public information (S) (S1030).[Pg 9, P-3]
Chung here illustrates providing at least part of the first ECG signal to a machine learning algorithm to obtain a first vector representation of first features of the first ECG signal, labelling the first vector representation as belonging to the authorized user, the labelling being based upon the first identity token
Chung also teaches obtain a second ECG signal for a first candidate user; provide at least part of the second ECG signal to the machine learning algorithm to obtain a second vector representation of second features of the second ECG signal; compare the second vector representation to the labelled first vector representation to determine the similarity metric; and if the similarity metric is within a predetermined threshold, providing confirmation that the first candidate user is the authorized user.(Abstract; Page 10, Paragraphs 5-11; Page 19, Paragraph 9-Page 20, Paragraph 7)
Disclosed are an identity verification apparatus for remotely measured biometric data and a method using the same. In an identity verification method according to an embodiment of the present invention, an identity verification device obtains a target ECG signal remotely measured for an authenticated user based on a wearable medical device, and based on a machine learning-based identity inference engine extracts at least one candidate user estimated to be a user of the target ECG signal from among a plurality of users registered in the authentication information database, and a reference feature vector of the at least one candidate user and a plurality of individual extracted from the target ECG signal The waveforms are compared, and the identity verification of the authenticated user is performed based on the identification result of each of the plurality of individual waveforms determined according to the comparison result. [Abstr]
Therefore, as a result of the determination in step S1055, when the error correction accuracy for the corrected target ECG signal is verified by matching the target verification hash value EV .sub.i '' and the verification hash value EV .sub.i , the correction target characteristic The vector W'' may be input to the identity inference engine to generate a candidate identity list (EID LIST) including at least one candidate user ( S1060 ). At this time, the corrected target feature vector (W'') by the successfully calibrated target ECG signal is input to the identity inference engine, and ID .sub.i <- Estimate(W'') based on Equation 6 is used to The user's identity can be inferred. For example, the identity inference engine identifies at least one candidate user (EID) having a predicted probability value as in [Equation 9], and generates a candidate identity list corresponding to the identified at least one candidate user (EID). Can [Pg 10, P-5-11]
In addition, the processor 1220 compares the reference feature vector of at least one candidate user and a plurality of individual waveforms extracted from the target ECG signal. In addition, the processor 1220 performs identity verification on the authenticated user based on the identification result for each of the plurality of individual waveforms determined according to the comparison result. That is, it is possible to verify whether the target ECG signal remotely measured for an authenticated user based on the wearable medical device comes from the patient's own body, which is expected to be monitored, and whether the person's identity is the same. Hereinafter, a process of verifying the source identity of the target ECG signal based on at least one candidate user (EID) estimated through Equation 9 will be described in detail with reference to FIG. 11. First, when a target electrocardiogram signal (ECG) remotely measured for a user (Real ID) authenticated based on a wearable medical device is received (S1110), at least one candidate as shown in Equation 9 using an identity inference engine The user (EID = {eID .sub.11 , ..., eID .sub.kn } may be extracted (S1120). In this case, the authenticated user's identifier (Real ID) may be obtained in a user authentication process for establishing a service session between a user using the wearable medical device and the identity verification device according to the present invention. In this case, the type of the identifier (Real ID) may not be limited. For example, a patient number that can be linked to a hospital's medical information system may be used as an identifier (Real ID). If a patient number that can be identified as a barcode or QR code can be extracted from the patient's ECG signal, an expected effect that can be utilized by easily linking with various hospital medical information systems linked to the patient number can occur. Thereafter, identity verification may be performed by comparing at least one candidate user (EID = {eID .sub.11 , ..., eID .sub.kn } with the authenticated user (Real ID) (S1130). In this case, a comparison check may be performed on both the authenticated user (Real ID) and at least one candidate user (eID .sub.ij ) identified in [Equation 9] until identity verification is successful. Thereafter, it is possible to perform a comparative test on each of the plurality of individual waveforms extracted from the target ECG signal, and determine whether identity verification is successful for all of the plurality of individual waveforms extracted from the target ECG signal ( S1135). In this case, it may be determined that the identification of the individual waveform matching the reference feature vector among the plurality of individual waveforms is successful.[Pg 19, P-9-Pg 20, P-7]
Here we see, Chung’s teaching after acquiring the token ID of a user affiliated with their ECG vector. The reference vector is stored in a database as the users ID. Thereafter, another candidate vector ID is acquired and compared accordingly to match the reference vector to determine if the user is an authorized user’s ID with respect to the reference vector.
Chung, however, fails to teach the authorization process being associated with, authorizing and verifying an authorized driver’s identity.
Zou on the other hand teaches obtaining a drivers biometric ID converted and stored in a database/server to which when a candidate vector ins read and compared to the stored authorized driver’s biometric vector ID, a match is determined, upon which, access to operate the vehicle is granted (Abstract; Page 2, Paragraph 5; Page 3, Paragraph 4)
The invention claims a vehicle-mounted system and operation method thereof. The vehicle-mounted system comprises a data obtaining device, a biological characteristic obtaining device and a processor. The data acquisition device is used for acquiring a self-key generated by de-identifying a first biological feature of a user using the vehicle to obtain first de-identified data and converting the first de-identified data into a first feature vector including a plurality of first de-identified features. The biological feature obtaining device is used for obtaining the second biological feature of the current user to be identified. The processor is used for de-identifying the second biological characteristic to obtain the second de-identified data, and converting it into the second characteristic vector containing multiple second de-identified characteristics, and comparing it with the first characteristic vector in the self key to start the pre-set function of the vehicle according to the comparison result.[Abstr]
The invention claims a vehicle system configured in a vehicle, comprising a data obtaining device, a first biological characteristic obtaining device and a processor. wherein the data obtaining device is used for obtaining the registered self-key, wherein the self-key is used for obtaining the first de-identification data by performing de-identification processing on the first biological characteristic of the user using the vehicle, and converting the first de-identified data into a first eigenvector containing a plurality of first de-identified features. The first biological characteristic obtaining device is used for obtaining the second biological characteristic of the current user to be identified. the processor is coupled with the data obtaining device and the first biological feature obtaining device, configured to perform de-identification processing on the second biological feature to obtain the second de-identification data, and convert the second de-identification data into a second feature vector comprising a plurality of second de-identification features, and comparing it with the first characteristic vector in the self key to start the predetermined function of the vehicle according to the comparison result.[Pg 2, P-5]
In some embodiments, the first biometric acquisition device is configured outside a vehicle for acquiring a third biometric of an external user. wherein the processor further performs de-identification processing on the third biological feature to obtain the third de-identification data, and converts the third de-identification data into the third feature vector containing a plurality of third de-identification features, and compares it with the first feature vector in its own key, The door of the vehicle is opened according to the comparison result.[Pg 3, P-4]
It would therefore be obvious during the time of the filing date to combine Zou’s teaching with Chung’s teaching in order to enable a more efficient and effective method to determine an authorized driver/operator of a vehicle, accordingly.
In regards to claim 20, Chung teaches a non-transitory computer-readable medium comprising instructions which, when executed by a computer, cause the computer to carry to: receive a first identity token from an authorized user; obtain, at the time of or after receiving the first identity token, a first electrocardiogram (ECG) signal for the authorized user(Abstract; Page 3, Paragraph 4)
Disclosed are an identity verification apparatus for remotely measured biometric data and a method using the same. In an identity verification method according to an embodiment of the present invention, an identity verification device obtains a target ECG signal remotely measured for an authenticated user based on a wearable medical device, and based on a machine learning-based identity inference engine extracts at least one candidate user estimated to be a user of the target ECG signal from among a plurality of users registered in the authentication information database, and a reference feature vector of the at least one candidate user and a plurality of individual extracted from the target ECG signal The waveforms are compared, and the identity verification of the authenticated user is performed based on the identification result of each of the plurality of individual waveforms determined according to the comparison result.[Abstr]
Therefore, the present invention was focused on the point that a patient's ECG signal has the characteristic of identifying an individual, and using this, a technical link that combines the ECG signal and the patient's identity can be made. That is, as shown in FIG. 2 , the identity of the source patient is identified using physiological data that changes with time and the measurement value is not constant at each measurement, and that the patient is the person corresponding to the monitoring target.[Pg 3. P-4]
For example, a new user visits a hospital face-to-face or non-face-to-face, measures his/her ECG signal based on his/her own or distributed Telehealth/Telemedicine service device and smartphone application, and registers it in the authentication information database, An identifier (ID) may be assigned corresponding to the patient number.[Pg 5, P-8]
Here, Chung discloses the computer implemented method receiving a first identity token from an authorized user, i.e. measures his/her ECG signal based on his/her own or distributed Telehealth/Telemedicine service device and smartphone application, and registers it in the authentication information database, An identifier (ID) may be assigned corresponding to the patient number. Furthermore, an identity verification device obtains a target ECG signal remotely measured for an authenticated user based on a wearable medical device, and based on a machine learning-based identity inference engine extracts at least one candidate user estimated to be a user of the target ECG signal from among a plurality of users registered in the authentication information database.
Chung then teaches provide at least part of the first ECG signal to a machine learning algorithm to obtain a first vector representation of first features of the first ECG signal; label the first vector representation as belonging to the authorized user, the labelling being based upon the first identity token(Abstract; Page 3, Paragraph 4)
Disclosed are an identity verification apparatus for remotely measured biometric data and a method using the same. In an identity verification method according to an embodiment of the present invention, an identity verification device obtains a target ECG signal remotely measured for an authenticated user based on a wearable medical device, and based on a machine learning-based identity inference engine extracts at least one candidate user estimated to be a user of the target ECG signal from among a plurality of users registered in the authentication information database, and a reference feature vector of the at least one candidate user and a plurality of individual extracted from the target ECG signal The waveforms are compared, and the identity verification of the authenticated user is performed based on the identification result of each of the plurality of individual waveforms determined according to the comparison result.[Abstr]
Therefore, the present invention was focused on the point that a patient's ECG signal has the characteristic of identifying an individual, and using this, a technical link that combines the ECG signal and the patient's identity can be made. That is, as shown in FIG. 2 , the identity of the source patient is identified using physiological data that changes with time and the measurement value is not constant at each measurement, and that the patient is the person corresponding to the monitoring target.[Pg 3. P-4]
For example, a new user visits a hospital face-to-face or non-face-to-face, measures his/her ECG signal based on his/her own or distributed Telehealth/Telemedicine service device and smartphone application, and registers it in the authentication information database, An identifier (ID) may be assigned corresponding to the patient number.[Pg 5, P-8]
First, when a target electrocardiogram signal (ECG) for a user authenticated based on a wearable medical device is received (S1010), a target feature vector (W') is extracted from the target electrocardiogram signal (S1020), and a target feature vector (W') ) can be input to the identity inference engine to derive target public information (S) (S1030).[Pg 9, P-3]
Furthermore, Chung teaches obtaining a second ECG signal for a first candidate user, providing at least part of the second ECG signal to the machine learning algorithm to obtain a second vector representation of second features of the second ECG signal, and compare the second vector representation to the labelled first vector representation to determine a similarity metric; and if the similarity metric is within a predetermined threshold, provide confirmation that the first candidate user is the authorized user(Abstract; Page 10, Paragraphs 5-11; Page 19, Paragraph 9-Page 20, Paragraph 7)
Disclosed are an identity verification apparatus for remotely measured biometric data and a method using the same. In an identity verification method according to an embodiment of the present invention, an identity verification device obtains a target ECG signal remotely measured for an authenticated user based on a wearable medical device, and based on a machine learning-based identity inference engine extracts at least one candidate user estimated to be a user of the target ECG signal from among a plurality of users registered in the authentication information database, and a reference feature vector of the at least one candidate user and a plurality of individual extracted from the target ECG signal The waveforms are compared, and the identity verification of the authenticated user is performed based on the identification result of each of the plurality of individual waveforms determined according to the comparison result. [Abstr]
Therefore, as a result of the determination in step S1055, when the error correction accuracy for the corrected target ECG signal is verified by matching the target verification hash value EV .sub.i '' and the verification hash value EV .sub.i , the correction target characteristic The vector W'' may be input to the identity inference engine to generate a candidate identity list (EID LIST) including at least one candidate user ( S1060 ). At this time, the corrected target feature vector (W'') by the successfully calibrated target ECG signal is input to the identity inference engine, and ID .sub.i <- Estimate(W'') based on Equation 6 is used to The user's identity can be inferred. For example, the identity inference engine identifies at least one candidate user (EID) having a predicted probability value as in [Equation 9], and generates a candidate identity list corresponding to the identified at least one candidate user (EID). Can [Pg 10, P-5-11]
In addition, the processor 1220 compares the reference feature vector of at least one candidate user and a plurality of individual waveforms extracted from the target ECG signal. In addition, the processor 1220 performs identity verification on the authenticated user based on the identification result for each of the plurality of individual waveforms determined according to the comparison result. That is, it is possible to verify whether the target ECG signal remotely measured for an authenticated user based on the wearable medical device comes from the patient's own body, which is expected to be monitored, and whether the person's identity is the same. Hereinafter, a process of verifying the source identity of the target ECG signal based on at least one candidate user (EID) estimated through Equation 9 will be described in detail with reference to FIG. 11. First, when a target electrocardiogram signal (ECG) remotely measured for a user (Real ID) authenticated based on a wearable medical device is received (S1110), at least one candidate as shown in Equation 9 using an identity inference engine The user (EID = {eID .sub.11 , ..., eID .sub.kn } may be extracted (S1120). In this case, the authenticated user's identifier (Real ID) may be obtained in a user authentication process for establishing a service session between a user using the wearable medical device and the identity verification device according to the present invention. In this case, the type of the identifier (Real ID) may not be limited. For example, a patient number that can be linked to a hospital's medical information system may be used as an identifier (Real ID). If a patient number that can be identified as a barcode or QR code can be extracted from the patient's ECG signal, an expected effect that can be utilized by easily linking with various hospital medical information systems linked to the patient number can occur. Thereafter, identity verification may be performed by comparing at least one candidate user (EID = {eID .sub.11 , ..., eID .sub.kn } with the authenticated user (Real ID) (S1130). In this case, a comparison check may be performed on both the authenticated user (Real ID) and at least one candidate user (eID .sub.ij ) identified in [Equation 9] until identity verification is successful. Thereafter, it is possible to perform a comparative test on each of the plurality of individual waveforms extracted from the target ECG signal, and determine whether identity verification is successful for all of the plurality of individual waveforms extracted from the target ECG signal ( S1135). In this case, it may be determined that the identification of the individual waveform matching the reference feature vector among the plurality of individual waveforms is successful.[Pg 19, P-9-Pg 20, P-7]
Here we see, Chung’s teaching after acquiring the token ID of a user affiliated with their ECG vector. The reference vector is stored in a database as the users ID. Thereafter, another candidate vector ID is acquired and compared accordingly to match the reference vector to determine if the user is an authorized user’s ID with respect to the reference vector.
Chung, however, fails to teach the authorization process being associated with, authorizing and verifying an authorized driver’s identity.
Zou on the other hand teaches obtaining a drivers biometric ID converted and stored in a database/server to which when a candidate vector ins read and compared to the stored authorized driver’s biometric vector ID, a match is determined, upon which, access to operate the vehicle is granted (Abstract; Page 2, Paragraph 5; Page 3, Paragraph 4)
The invention claims a vehicle-mounted system and operation method thereof. The vehicle-mounted system comprises a data obtaining device, a biological characteristic obtaining device and a processor. The data acquisition device is used for acquiring a self-key generated by de-identifying a first biological feature of a user using the vehicle to obtain first de-identified data and converting the first de-identified data into a first feature vector including a plurality of first de-identified features. The biological feature obtaining device is used for obtaining the second biological feature of the current user to be identified. The processor is used for de-identifying the second biological characteristic to obtain the second de-identified data, and converting it into the second characteristic vector containing multiple second de-identified characteristics, and comparing it with the first characteristic vector in the self key to start the pre-set function of the vehicle according to the comparison result.[Abstr]
The invention claims a vehicle system configured in a vehicle, comprising a data obtaining device, a first biological characteristic obtaining device and a processor. wherein the data obtaining device is used for obtaining the registered self-key, wherein the self-key is used for obtaining the first de-identification data by performing de-identification processing on the first biological characteristic of the user using the vehicle, and converting the first de-identified data into a first eigenvector containing a plurality of first de-identified features. The first biological characteristic obtaining device is used for obtaining the second biological characteristic of the current user to be identified. the processor is coupled with the data obtaining device and the first biological feature obtaining device, configured to perform de-identification processing on the second biological feature to obtain the second de-identification data, and convert the second de-identification data into a second feature vector comprising a plurality of second de-identification features, and comparing it with the first characteristic vector in the self key to start the predetermined function of the vehicle according to the comparison result.[Pg 2, P-5]
In some embodiments, the first biometric acquisition device is configured outside a vehicle for acquiring a third biometric of an external user. wherein the processor further performs de-identification processing on the third biological feature to obtain the third de-identification data, and converts the third de-identification data into the third feature vector containing a plurality of third de-identification features, and compares it with the first feature vector in its own key, The door of the vehicle is opened according to the comparison result.[Pg 3, P-4]
It would therefore be obvious during the time of the filing date to combine Zou’s teaching with Chung’s teaching in order to enable a more efficient and effective method to determine an authorized driver/operator of a vehicle, accordingly.
Claim(s) 2, 6, 14, 21, 25-27, and 31 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chung et al. (KR 20220077495 A) in views Zou (CN 118041543 A) as applied to claim 1, 15, 20, above, and further in view of Zhu et al. (CN 112494044 A).
In regards to claim 2, Chung modified fails to teach the first ECG signal is obtained for a longer time duration than the second ECG signal and comprises a plurality of ECG signal segments taken at different times
Zhu on the other hand teaches the first ECG signal is obtained for a longer time duration than the second ECG signal and comprises a plurality of ECG signal segments taken at different times (Page 6, Last Paragraph – Page 7, Paragraph 1)
electrocardiogram signal time period is a continuous signal; it is not convenient for subsequent data analysis; therefore, in the present disclosure, after obtaining the electrocardiogram signal in different time period, sampling for each time period of the electrocardiogram signal to obtain the discrete electrocardiogram data; and these ECG data can form an ECG vector. wherein the duration of each time period are the same, which can be 2s, 5s and so on, the present disclosure is not specifically limited.[Pg 6, Lst Pgh- Pg 7, P-1]
Furthermore, Zhu teaches each of the plurality of ECG signal segments of the first ECG signal is provided to the machine learning algorithm individually so as to produce a plurality of first vector representations, and wherein the computer implemented method further comprises combining the plurality of first vector representations to form a first vector representation space that is indicative of the first features of the plurality of ECG signal segments.(Page 7, Paragraphs 3-7)
in the step 103, clustering the plurality of ECG vector by the preset clustering algorithm to obtain the effective electrocardiogram vector. Because there are various unstable vibration frequency in the vehicle environment, the electrocardiogram signal have serious distortion phenomenon, the electrocardiogram signal distorted cannot reliably reflect the health state or mental state of the driver; therefore, in the present disclosure, before performing electrocardiogram analysis to the ECG vector; screening the plurality of ECG vectors, to screen out the effective ECG vector, eliminating the invalid ECG vector. example, the plurality of ECG vectors can be clustered by clustering algorithm to distinguish the effective ECG vector and invalid ECG vector. It should be noted that the invalid ECG vector is an electrocardiogram signal group.[Pg 7, P-3-4]
in the step 104, identifying the key point of the electrocardiogram signal electrocardiogram vector. the key point of the electrocardiogram signal is capable of characterizing the electrocardiogram signal characteristic, and convenient for subsequent analysis of the driver health state or mental state of the sampling point, the key point generally comprises a key point of P, Q, R, S, T and so on corresponding to the electrocardiogram signal wave pattern. example, the key point may include at least one of the following: P wave peak point, R wave position point, T wave position point, Q wave position point and S wave position point. example, it can pass through the WFDB electrocardiogram signal tool bag, firstly, determining R wave in the effective electrocardiogram vector, then respectively forward detecting P wave starting point (P-onset), P-wave (P-wave), P-wave termination point (P-offset) on the basis of R-wave; Q-wave (Q-wave), S-wave (S-wave), T-wave (T_wave) and T-wave termination point (T_offset). after determining the key points, combining the collected original electrocardiogram signal, obtaining the electrocardiogram amplitude corresponding to each key point. For example, Table 1 is the electrocardiogram amplitude corresponding to each key point and key point detected from each electrocardiogram vector.[Pg 7, P-5-7]
We see, Zhu illustrate combining the plurality of first vector representations to form a first vector representation space that is indicative of the first features of the plurality of ECG signal segments, i.e. determining R wave in the effective electrocardiogram vector, then respectively forward detecting P wave starting point (P-onset), P-wave (P-wave), P-wave termination point (P-offset) on the basis of R-wave; Q-wave (Q-wave), S-wave (S-wave), T-wave (T_wave) and T-wave termination point (T_offset). after determining the key points, combining the collected original electrocardiogram signal
It would therefore be obvious during the time of the filing date to combine Zhu’s teaching with Chung modified’s teaching in order to acquire and generate a vector that represent ECG signal(s) accordingly.
In regards to claim 6, Chung modified teaches the second ECG signal comprises a plurality of ECG signal segments taken at different times, wherein each of the plurality of ECG signal segments is provided to the machine learning algorithm individually so as to produce a plurality of second vector representations, and wherein the computer implemented method further comprises combining the plurality of second vector representations to form a second vector representation space that is indicative of the second features of the plurality of ECG signal segments.
Zhu on the other hand teaches the second ECG signal comprises a plurality of ECG signal segments taken at different times, wherein each of the plurality of ECG signal segments is provided to the machine learning algorithm individually so as to produce a plurality of second vector representations, and wherein the computer implemented method further comprises combining the plurality of second vector representations to form a second vector representation space that is indicative of the second features of the plurality of ECG signal segments. (Page 3, Paragraphs 9-10; Page 6, Last Paragraph – Page 7, Paragraph 1; Page 7, Paragraphs 3-7; Page 10, Paragraphs 5-6)
according to the correlation algorithm, determining the correlation coefficient of each of the second effective electrocardiogram vector and the sample ECG vector; The second effective ECG vector with the correlation coefficient greater than the correlation coefficient threshold is determined as an effective electrocardiogram signal[Pg3, P-9-10]
electrocardiogram signal time period is a continuous signal; it is not convenient for subsequent data analysis; therefore, in the present disclosure, after obtaining the electrocardiogram signal in different time period, sampling for each time period of the electrocardiogram signal to obtain the discrete electrocardiogram data; and these ECG data can form an ECG vector. wherein the duration of each time period are the same, which can be 2s, 5s and so on, the present disclosure is not specifically limited.[Pg 6, Lst Pgh- Pg 7, P-1]
in the step 103, clustering the plurality of ECG vector by the preset clustering algorithm to obtain the effective electrocardiogram vector. Because there are various unstable vibration frequency in the vehicle environment, the electrocardiogram signal have serious distortion phenomenon, the electrocardiogram signal distorted cannot reliably reflect the health state or mental state of the driver; therefore, in the present disclosure, before performing electrocardiogram analysis to the ECG vector; screening the plurality of ECG vectors, to screen out the effective ECG vector, eliminating the invalid ECG vector. example, the plurality of ECG vectors can be clustered by clustering algorithm to distinguish the effective ECG vector and invalid ECG vector. It should be noted that the invalid ECG vector is an electrocardiogram signal group.[Pg 7, P-3-4]
in the step 104, identifying the key point of the electrocardiogram signal electrocardiogram vector. the key point of the electrocardiogram signal is capable of characterizing the electrocardiogram signal characteristic, and convenient for subsequent analysis of the driver health state or mental state of the sampling point, the key point generally comprises a key point of P, Q, R, S, T and so on corresponding to the electrocardiogram signal wave pattern. example, the key point may include at least one of the following: P wave peak point, R wave position point, T wave position point, Q wave position point and S wave position point. example, it can pass through the WFDB electrocardiogram signal tool bag, firstly, determining R wave in the effective electrocardiogram vector, then respectively forward detecting P wave starting point (P-onset), P-wave (P-wave), P-wave termination point (P-offset) on the basis of R-wave; Q-wave (Q-wave), S-wave (S-wave), T-wave (T_wave) and T-wave termination point (T_offset). after determining the key points, combining the collected original electrocardiogram signal, obtaining the electrocardiogram amplitude corresponding to each key point. For example, Table 1 is the electrocardiogram amplitude corresponding to each key point and key point detected from each electrocardiogram vector.[Pg 7, P-5-7]
then, according to the correlation algorithm, determining the correlation coefficient of each second effective electrocardiogram vector and sample ECG vector, and determining the second effective electrocardiogram vector with correlation coefficient greater than the correlation coefficient threshold as effective electrocardiogram signal Similarly, only calculating the cosine similarity is greater than cosine similarity threshold value of the second effective electrocardiogram vector and the correlation coefficient of the sample electrocardiogram vector, further reduces the workload of calculating the correlation coefficient. wherein, can refer to the formula (5) to calculate the correlation coefficient of each second effective electrocardiogram vector and sample electrocardiogram vector, here.[Pg 10, P-5-6]
Here, Zhu discloses a second ECG vector representative of an second ECG signal. Furthermore, Zhu discloses, the plurality of ECG signal segments is provided to the machine learning algorithm individually so as to produce a plurality of vector representations, and wherein the computer implemented method further comprises combining the plurality of vector representations to form a vector representation space that is indicative of the features of the plurality of ECG signal segments. Hence, the said procedure being applicable to the second ECG vector.
It would therefore be obvious during the time of the filing date to combine Zhu’s teaching with Chung modified’s teaching in order to acquire and generate a vector that represent ECG signal(s) accordingly.
In regards to claim 14, Chung modified fails to teach comparing the second vector representation to the labelled first vector representation comprises determining a distance in a vector space between the second vector representation and the labelled first vector representation, and wherein the similarity metric is a measure of the distance.
Zhu on the other hand teaches comparing the second vector representation to the labelled first vector representation comprises determining a distance in a vector space between the second vector representation and the labelled first vector representation, and wherein the similarity metric is a measure of the distance.(Page 3, Paragraphs 4-6; Claim 4)
Optionally, the preset clustering algorithm comprises a Mahalanobis distance algorithm and similarity algorithm, the similarity algorithm comprises a cosine similarity algorithm and correlation algorithm; the plurality of the ECG vector is clustered by the preset clustering algorithm to obtain the effective ECG vector, comprising: through the Mahalanobis distance algorithm, determining the Mahalanobis distance of each electrocardiogram vector and the Mahalanobis distance central point of a plurality of the electrocardiogram vector; determining the electrocardiograph of the Mahalanobis distance in the preset confidence interval as the first effective electrocardiograph [Pg3 , P-4-6]
The method according to claim 1, wherein the preset clustering algorithm comprises a Mahalanobis distance algorithm and a similarity algorithm; the similarity algorithm comprises a cosine similarity algorithm and a correlation algorithm; the plurality of the ECG vector is clustered by the preset clustering algorithm to obtain the effective ECG vector, comprising: through the Mahalanobis distance algorithm, determining the Mahalanobis distance of each electrocardiogram vector and the Mahalanobis distance central point of a plurality of the electrocardiogram vector; determining the electrocardiograph of the Mahalanobis distance in the preset confidence interval as the first effective electrocardiograph; according to the cosine similarity algorithm, determining cosine similarity of each of the first effective electrocardiogram vector and sample electrocardiogram vector; determining the first effective ECG vector with the cosine similarity greater than the cosine similarity threshold as the second effective ECG vector; according to the correlation algorithm, determining the correlation coefficient of each of the second effective electrocardiogram vector and the sample ECG vector; The second effective ECG vector with the correlation coefficient greater than the correlation coefficient threshold is determined as an effective electrocardiogram signal[Cl-4]
It would therefore be obvious during the time of the filing date to combine Zhu’s teaching with Chung modified’s teaching in order to determine the electrocardiograph of the distance in the preset confidence interval as the effective ECG accordingly.
In regards to claim 21, Chung modified fails to teach the first ECG signal is obtained for a longer time duration than the second ECG signal and comprises a plurality of ECG signal segments taken at different times
Zhu on the other hand teaches the first ECG signal is obtained for a longer time duration than the second ECG signal and comprises a plurality of ECG signal segments taken at different times (Page 6, Last Paragraph – Page 7, Paragraph 1)
electrocardiogram signal time period is a continuous signal; it is not convenient for subsequent data analysis; therefore, in the present disclosure, after obtaining the electrocardiogram signal in different time period, sampling for each time period of the electrocardiogram signal to obtain the discrete electrocardiogram data; and these ECG data can form an ECG vector. wherein the duration of each time period are the same, which can be 2s, 5s and so on, the present disclosure is not specifically limited.[Pg 6, Lst Pgh- Pg 7, P-1]
Furthermore, Zhu teaches each of the plurality of ECG signal segments of the first ECG signal is provided to the machine learning algorithm individually so as to produce a plurality of first vector representations, and wherein the computer implemented method further comprises combining the plurality of first vector representations to form a first vector representation space that is indicative of the first features of the plurality of ECG signal segments.(Page 7, Paragraphs 3-7)
in the step 103, clustering the plurality of ECG vector by the preset clustering algorithm to obtain the effective electrocardiogram vector. Because there are various unstable vibration frequency in the vehicle environment, the electrocardiogram signal have serious distortion phenomenon, the electrocardiogram signal distorted cannot reliably reflect the health state or mental state of the driver; therefore, in the present disclosure, before performing electrocardiogram analysis to the ECG vector; screening the plurality of ECG vectors, to screen out the effective ECG vector, eliminating the invalid ECG vector. example, the plurality of ECG vectors can be clustered by clustering algorithm to distinguish the effective ECG vector and invalid ECG vector. It should be noted that the invalid ECG vector is an electrocardiogram signal group.[Pg 7, P-3-4]
in the step 104, identifying the key point of the electrocardiogram signal electrocardiogram vector. the key point of the electrocardiogram signal is capable of characterizing the electrocardiogram signal characteristic, and convenient for subsequent analysis of the driver health state or mental state of the sampling point, the key point generally comprises a key point of P, Q, R, S, T and so on corresponding to the electrocardiogram signal wave pattern. example, the key point may include at least one of the following: P wave peak point, R wave position point, T wave position point, Q wave position point and S wave position point. example, it can pass through the WFDB electrocardiogram signal tool bag, firstly, determining R wave in the effective electrocardiogram vector, then respectively forward detecting P wave starting point (P-onset), P-wave (P-wave), P-wave termination point (P-offset) on the basis of R-wave; Q-wave (Q-wave), S-wave (S-wave), T-wave (T_wave) and T-wave termination point (T_offset). after determining the key points, combining the collected original electrocardiogram signal, obtaining the electrocardiogram amplitude corresponding to each key point. For example, Table 1 is the electrocardiogram amplitude corresponding to each key point and key point detected from each electrocardiogram vector.[Pg 7, P-5-7]
We see, Zhu illustrate combining the plurality of first vector representations to form a first vector representation space that is indicative of the first features of the plurality of ECG signal segments, i.e. determining R wave in the effective electrocardiogram vector, then respectively forward detecting P wave starting point (P-onset), P-wave (P-wave), P-wave termination point (P-offset) on the basis of R-wave; Q-wave (Q-wave), S-wave (S-wave), T-wave (T_wave) and T-wave termination point (T_offset). after determining the key points, combining the collected original electrocardiogram signal
It would therefore be obvious during the time of the filing date to combine Zhu’s teaching with Chung modified’s teaching in order to acquire and generate a vector that represent ECG signal(s) accordingly.
In regards to claim 25, Chung modified teaches the second ECG signal comprises a plurality of ECG signal segments taken at different times, wherein each of the plurality of ECG signal segments is provided to the machine learning algorithm individually so as to produce a plurality of second vector representations, and wherein the computer implemented method further comprises combining the plurality of second vector representations to form a second vector representation space that is indicative of the second features of the plurality of ECG signal segments.
Zhu on the other hand teaches the second ECG signal comprises a plurality of ECG signal segments taken at different times, wherein each of the plurality of ECG signal segments is provided to the machine learning algorithm individually so as to produce a plurality of second vector representations, and wherein the computer implemented method further comprises combining the plurality of second vector representations to form a second vector representation space that is indicative of the second features of the plurality of ECG signal segments. (Page 3, Paragraphs 9-10; Page 6, Last Paragraph – Page 7, Paragraph 1; Page 7, Paragraphs 3-7; Page 10, Paragraphs 5-6)
according to the correlation algorithm, determining the correlation coefficient of each of the second effective electrocardiogram vector and the sample ECG vector; The second effective ECG vector with the correlation coefficient greater than the correlation coefficient threshold is determined as an effective electrocardiogram signal[Pg3, P-9-10]
electrocardiogram signal time period is a continuous signal; it is not convenient for subsequent data analysis; therefore, in the present disclosure, after obtaining the electrocardiogram signal in different time period, sampling for each time period of the electrocardiogram signal to obtain the discrete electrocardiogram data; and these ECG data can form an ECG vector. wherein the duration of each time period are the same, which can be 2s, 5s and so on, the present disclosure is not specifically limited.[Pg 6, Lst Pgh- Pg 7, P-1]
in the step 103, clustering the plurality of ECG vector by the preset clustering algorithm to obtain the effective electrocardiogram vector. Because there are various unstable vibration frequency in the vehicle environment, the electrocardiogram signal have serious distortion phenomenon, the electrocardiogram signal distorted cannot reliably reflect the health state or mental state of the driver; therefore, in the present disclosure, before performing electrocardiogram analysis to the ECG vector; screening the plurality of ECG vectors, to screen out the effective ECG vector, eliminating the invalid ECG vector. example, the plurality of ECG vectors can be clustered by clustering algorithm to distinguish the effective ECG vector and invalid ECG vector. It should be noted that the invalid ECG vector is an electrocardiogram signal group.[Pg 7, P-3-4]
in the step 104, identifying the key point of the electrocardiogram signal electrocardiogram vector. the key point of the electrocardiogram signal is capable of characterizing the electrocardiogram signal characteristic, and convenient for subsequent analysis of the driver health state or mental state of the sampling point, the key point generally comprises a key point of P, Q, R, S, T and so on corresponding to the electrocardiogram signal wave pattern. example, the key point may include at least one of the following: P wave peak point, R wave position point, T wave position point, Q wave position point and S wave position point. example, it can pass through the WFDB electrocardiogram signal tool bag, firstly, determining R wave in the effective electrocardiogram vector, then respectively forward detecting P wave starting point (P-onset), P-wave (P-wave), P-wave termination point (P-offset) on the basis of R-wave; Q-wave (Q-wave), S-wave (S-wave), T-wave (T_wave) and T-wave termination point (T_offset). after determining the key points, combining the collected original electrocardiogram signal, obtaining the electrocardiogram amplitude corresponding to each key point. For example, Table 1 is the electrocardiogram amplitude corresponding to each key point and key point detected from each electrocardiogram vector.[Pg 7, P-5-7]
then, according to the correlation algorithm, determining the correlation coefficient of each second effective electrocardiogram vector and sample ECG vector, and determining the second effective electrocardiogram vector with correlation coefficient greater than the correlation coefficient threshold as effective electrocardiogram signal Similarly, only calculating the cosine similarity is greater than cosine similarity threshold value of the second effective electrocardiogram vector and the correlation coefficient of the sample electrocardiogram vector, further reduces the workload of calculating the correlation coefficient. wherein, can refer to the formula (5) to calculate the correlation coefficient of each second effective electrocardiogram vector and sample electrocardiogram vector, here.[Pg 10, P-5-6]
Here, Zhu discloses a second ECG vector representative of an second ECG signal. Furthermore, Zhu discloses, the plurality of ECG signal segments is provided to the machine learning algorithm individually so as to produce a plurality of vector representations, and wherein the computer implemented method further comprises combining the plurality of vector representations to form a vector representation space that is indicative of the features of the plurality of ECG signal segments. Hence, the said procedure being applicable to the second ECG vector.
It would therefore be obvious during the time of the filing date to combine Zhu’s teaching with Chung modified’s teaching in order to acquire and generate a vector that represent ECG signal(s) accordingly.
In regards to claim 26, Chung modified fails to teach comparing the second vector representation to the labelled first vector representation comprises determining a distance in a vector space between the second vector representation and the labelled first vector representation, and wherein the similarity metric is a measure of the distance.
Zhu on the other hand teaches comparing the second vector representation to the labelled first vector representation comprises determining a distance in a vector space between the second vector representation and the labelled first vector representation, and wherein the similarity metric is a measure of the distance.(Page 3, Paragraphs 4-6; Claim 4)
Optionally, the preset clustering algorithm comprises a Mahalanobis distance algorithm and similarity algorithm, the similarity algorithm comprises a cosine similarity algorithm and correlation algorithm; the plurality of the ECG vector is clustered by the preset clustering algorithm to obtain the effective ECG vector, comprising: through the Mahalanobis distance algorithm, determining the Mahalanobis distance of each electrocardiogram vector and the Mahalanobis distance central point of a plurality of the electrocardiogram vector; determining the electrocardiograph of the Mahalanobis distance in the preset confidence interval as the first effective electrocardiograph [Pg3 , P-4-6]
The method according to claim 1, wherein the preset clustering algorithm comprises a Mahalanobis distance algorithm and a similarity algorithm; the similarity algorithm comprises a cosine similarity algorithm and a correlation algorithm; the plurality of the ECG vector is clustered by the preset clustering algorithm to obtain the effective ECG vector, comprising: through the Mahalanobis distance algorithm, determining the Mahalanobis distance of each electrocardiogram vector and the Mahalanobis distance central point of a plurality of the electrocardiogram vector; determining the electrocardiograph of the Mahalanobis distance in the preset confidence interval as the first effective electrocardiograph; according to the cosine similarity algorithm, determining cosine similarity of each of the first effective electrocardiogram vector and sample electrocardiogram vector; determining the first effective ECG vector with the cosine similarity greater than the cosine similarity threshold as the second effective ECG vector; according to the correlation algorithm, determining the correlation coefficient of each of the second effective electrocardiogram vector and the sample ECG vector; The second effective ECG vector with the correlation coefficient greater than the correlation coefficient threshold is determined as an effective electrocardiogram signal[Cl-4]
It would therefore be obvious during the time of the filing date to combine Zhu’s teaching with Chung modified’s teaching in order to determine the electrocardiograph of the distance in the preset confidence interval as the effective ECG accordingly.
In regards to claim 27, Chung modified fails to teach the first ECG signal is obtained for a longer time duration than the second ECG signal and comprises a plurality of ECG signal segments taken at different times
Zhu on the other hand teaches the first ECG signal is obtained for a longer time duration than the second ECG signal and comprises a plurality of ECG signal segments taken at different times (Page 6, Last Paragraph – Page 7, Paragraph 1)
electrocardiogram signal time period is a continuous signal; it is not convenient for subsequent data analysis; therefore, in the present disclosure, after obtaining the electrocardiogram signal in different time period, sampling for each time period of the electrocardiogram signal to obtain the discrete electrocardiogram data; and these ECG data can form an ECG vector. wherein the duration of each time period are the same, which can be 2s, 5s and so on, the present disclosure is not specifically limited.[Pg 6, Lst Pgh- Pg 7, P-1]
Furthermore, Zhu teaches each of the plurality of ECG signal segments of the first ECG signal is provided to the machine learning algorithm individually so as to produce a plurality of first vector representations, and wherein the computer implemented method further comprises combining the plurality of first vector representations to form a first vector representation space that is indicative of the first features of the plurality of ECG signal segments.(Page 7, Paragraphs 3-7)
in the step 103, clustering the plurality of ECG vector by the preset clustering algorithm to obtain the effective electrocardiogram vector. Because there are various unstable vibration frequency in the vehicle environment, the electrocardiogram signal have serious distortion phenomenon, the electrocardiogram signal distorted cannot reliably reflect the health state or mental state of the driver; therefore, in the present disclosure, before performing electrocardiogram analysis to the ECG vector; screening the plurality of ECG vectors, to screen out the effective ECG vector, eliminating the invalid ECG vector. example, the plurality of ECG vectors can be clustered by clustering algorithm to distinguish the effective ECG vector and invalid ECG vector. It should be noted that the invalid ECG vector is an electrocardiogram signal group.[Pg 7, P-3-4]
in the step 104, identifying the key point of the electrocardiogram signal electrocardiogram vector. the key point of the electrocardiogram signal is capable of characterizing the electrocardiogram signal characteristic, and convenient for subsequent analysis of the driver health state or mental state of the sampling point, the key point generally comprises a key point of P, Q, R, S, T and so on corresponding to the electrocardiogram signal wave pattern. example, the key point may include at least one of the following: P wave peak point, R wave position point, T wave position point, Q wave position point and S wave position point. example, it can pass through the WFDB electrocardiogram signal tool bag, firstly, determining R wave in the effective electrocardiogram vector, then respectively forward detecting P wave starting point (P-onset), P-wave (P-wave), P-wave termination point (P-offset) on the basis of R-wave; Q-wave (Q-wave), S-wave (S-wave), T-wave (T_wave) and T-wave termination point (T_offset). after determining the key points, combining the collected original electrocardiogram signal, obtaining the electrocardiogram amplitude corresponding to each key point. For example, Table 1 is the electrocardiogram amplitude corresponding to each key point and key point detected from each electrocardiogram vector.[Pg 7, P-5-7]
We see, Zhu illustrate combining the plurality of first vector representations to form a first vector representation space that is indicative of the first features of the plurality of ECG signal segments, i.e. determining R wave in the effective electrocardiogram vector, then respectively forward detecting P wave starting point (P-onset), P-wave (P-wave), P-wave termination point (P-offset) on the basis of R-wave; Q-wave (Q-wave), S-wave (S-wave), T-wave (T_wave) and T-wave termination point (T_offset). after determining the key points, combining the collected original electrocardiogram signal
It would therefore be obvious during the time of the filing date to combine Zhu’s teaching with Chung modified’s teaching in order to acquire and generate a vector that represent ECG signal(s) accordingly.
In regards to claim 31, Chung modified teaches the second ECG signal comprises a plurality of ECG signal segments taken at different times, wherein each of the plurality of ECG signal segments is provided to the machine learning algorithm individually so as to produce a plurality of second vector representations, and wherein the computer implemented method further comprises combining the plurality of second vector representations to form a second vector representation space that is indicative of the second features of the plurality of ECG signal segments.
Zhu on the other hand teaches the second ECG signal comprises a plurality of ECG signal segments taken at different times, wherein each of the plurality of ECG signal segments is provided to the machine learning algorithm individually so as to produce a plurality of second vector representations, and wherein the computer implemented method further comprises combining the plurality of second vector representations to form a second vector representation space that is indicative of the second features of the plurality of ECG signal segments. (Page 3, Paragraphs 9-10; Page 6, Last Paragraph – Page 7, Paragraph 1; Page 7, Paragraphs 3-7; Page 10, Paragraphs 5-6)
according to the correlation algorithm, determining the correlation coefficient of each of the second effective electrocardiogram vector and the sample ECG vector; The second effective ECG vector with the correlation coefficient greater than the correlation coefficient threshold is determined as an effective electrocardiogram signal[Pg3, P-9-10]
electrocardiogram signal time period is a continuous signal; it is not convenient for subsequent data analysis; therefore, in the present disclosure, after obtaining the electrocardiogram signal in different time period, sampling for each time period of the electrocardiogram signal to obtain the discrete electrocardiogram data; and these ECG data can form an ECG vector. wherein the duration of each time period are the same, which can be 2s, 5s and so on, the present disclosure is not specifically limited.[Pg 6, Lst Pgh- Pg 7, P-1]
in the step 103, clustering the plurality of ECG vector by the preset clustering algorithm to obtain the effective electrocardiogram vector. Because there are various unstable vibration frequency in the vehicle environment, the electrocardiogram signal have serious distortion phenomenon, the electrocardiogram signal distorted cannot reliably reflect the health state or mental state of the driver; therefore, in the present disclosure, before performing electrocardiogram analysis to the ECG vector; screening the plurality of ECG vectors, to screen out the effective ECG vector, eliminating the invalid ECG vector. example, the plurality of ECG vectors can be clustered by clustering algorithm to distinguish the effective ECG vector and invalid ECG vector. It should be noted that the invalid ECG vector is an electrocardiogram signal group.[Pg 7, P-3-4]
in the step 104, identifying the key point of the electrocardiogram signal electrocardiogram vector. the key point of the electrocardiogram signal is capable of characterizing the electrocardiogram signal characteristic, and convenient for subsequent analysis of the driver health state or mental state of the sampling point, the key point generally comprises a key point of P, Q, R, S, T and so on corresponding to the electrocardiogram signal wave pattern. example, the key point may include at least one of the following: P wave peak point, R wave position point, T wave position point, Q wave position point and S wave position point. example, it can pass through the WFDB electrocardiogram signal tool bag, firstly, determining R wave in the effective electrocardiogram vector, then respectively forward detecting P wave starting point (P-onset), P-wave (P-wave), P-wave termination point (P-offset) on the basis of R-wave; Q-wave (Q-wave), S-wave (S-wave), T-wave (T_wave) and T-wave termination point (T_offset). after determining the key points, combining the collected original electrocardiogram signal, obtaining the electrocardiogram amplitude corresponding to each key point. For example, Table 1 is the electrocardiogram amplitude corresponding to each key point and key point detected from each electrocardiogram vector.[Pg 7, P-5-7]
then, according to the correlation algorithm, determining the correlation coefficient of each second effective electrocardiogram vector and sample ECG vector, and determining the second effective electrocardiogram vector with correlation coefficient greater than the correlation coefficient threshold as effective electrocardiogram signal Similarly, only calculating the cosine similarity is greater than cosine similarity threshold value of the second effective electrocardiogram vector and the correlation coefficient of the sample electrocardiogram vector, further reduces the workload of calculating the correlation coefficient. wherein, can refer to the formula (5) to calculate the correlation coefficient of each second effective electrocardiogram vector and sample electrocardiogram vector, here.[Pg 10, P-5-6]
Here, Zhu discloses a second ECG vector representative of an second ECG signal. Furthermore, Zhu discloses, the plurality of ECG signal segments is provided to the machine learning algorithm individually so as to produce a plurality of vector representations, and wherein the computer implemented method further comprises combining the plurality of vector representations to form a vector representation space that is indicative of the features of the plurality of ECG signal segments. Hence, the said procedure being applicable to the second ECG vector.
It would therefore be obvious during the time of the filing date to combine Zhu’s teaching with Chung modified’s teaching in order to acquire and generate a vector that represent ECG signal(s) accordingly.
Allowable Subject Matter
Claims 3-5, 22-24, 28-30 are 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.
Claim 3, reads as “The computer implemented method of claim 2, wherein if the similarity metric is outside of the predetermined threshold, the computer implemented method further comprises: retrieving a second identity token from the first candidate driver prior to or at the time of obtaining the second ECG signal; comparing the second identity token to the first identity token provided by the authorized driver; and if the second identity token corresponds to the first identity token, combining the second vector representation with the plurality of first vector representations, thereby updating the first vector representation space.”
Chung et al. (KR 20220077495 A), Zou (CN 118041543 A), and Zhu et al. (CN 112494044 A), individually or in combination fail to teach the retrieving a second identity token from the first candidate driver prior to or at the time of obtaining the second ECG signal; comparing the second identity token to the first identity token, such that the second identity token corresponds to the first identity token, combining the second vector representation with the plurality of first vector representations, thereby updating the first vector representation space. Dependent claims 4 and 5 are objected for the same reason.
Claim 22, reads as “The non-transitory computer-readable medium of claim 21, wherein if the similarity metric is outside of the predetermined threshold, the instructions, when executed by the computer, further cause the computer to: retrieve a second identity token from the first candidate driver prior to or at the time of obtaining the second ECG signal; compare the second identity token to the first identity token provided by the authorized driver; and if the second identity token corresponds to the first identity token, combine the second vector representation with the plurality of first vector representations, thereby updating the first vector representation space.”
Chung et al. (KR 20220077495 A), Zou (CN 118041543 A), and Zhu et al. (CN 112494044 A), individually or in combination fail to teach the retrieving a second identity token from the first candidate driver prior to or at the time of obtaining the second ECG signal; comparing the second identity token to the first identity token, such that the second identity token corresponds to the first identity token, combining the second vector representation with the plurality of first vector representations, thereby updating the first vector representation space. Dependent claims 23 and 24 are objected for the same reason.
Claim 28, reads as “The system of claim 27, wherein if the similarity metric is outside of the predetermined threshold, the system is further configured to: retrieve a second identity token from the first candidate driver prior to or at the time of obtaining the second ECG signal; compare the second identity token to the first identity token provided by the authorized driver; and if the second identity token corresponds to the first identity token, combine the second vector representation with the plurality of first vector representations, thereby updating the first vector representation space.”
Chung et al. (KR 20220077495 A), Zou (CN 118041543 A), and Zhu et al. (CN 112494044 A), individually or in combination fail to teach the retrieving a second identity token from the first candidate driver prior to or at the time of obtaining the second ECG signal; comparing the second identity token to the first identity token, such that the second identity token corresponds to the first identity token, combining the second vector representation with the plurality of first vector representations, thereby updating the first vector representation space. Dependent claims 29 and 30 are objected for the same reason.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to ANTHONY D AFRIFA-KYEI whose telephone number is (571)270-7826. The examiner can normally be reached Monday-Friday 10am-7pm.
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, BRIAN ZIMMERMAN can be reached at 571-272-3059. 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.
/ANTHONY D AFRIFA-KYEI/Examiner, Art Unit 2686
/BRIAN A ZIMMERMAN/Supervisory Patent Examiner, Art Unit 2686
