# US 2021/0330203 A1 · Blood pressure measuring apparatus, blood pressure measuring method, electronic device, and computer readable storage medium > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2021/0330203 A1 - Document ID: US20210330203A1 - Status: Published application - Kind: A1 - Application number: 16/489,696 - Filed: 2019-01-31 - Published: 2021-10-28 - Earliest claimed: 2018-07-03 - Assignees: Boe Technology Group Co., Ltd. - Applicants: Boe Technology Group Co., Ltd. - Inventors: Zhihong Xu - Web page: https://patentgrep.com/patent/US20210330203A1 ## CPC - A61B 5/02125: {of pulse wave propagation time} - A61B 5/0205: Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition - A61B 5/14542: {for measuring blood gases (A61B5/14551 takes precedence)} - A61B 5/14551: {for measuring blood gases} - A61B 5/318: Heart-related electrical modalities, e.g. electrocardiography [ECG] - A61B 5/7267: {involving training the classification device} - A61B 2560/02: Operational features - A61B 2560/0223: of calibration, e.g. protocols for calibrating sensors - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) ## Abstract Embodiments of the present disclosure provide a blood pressure measuring method, a blood pressure measuring apparatus, an electronic device and a computer readable storage medium. The electronic device includes a processor configured to: acquire an electrocardiosignal and a blood oxygen volume wave signal of a target object within a preset measurement period; and determine beat-wise blood pressure values of the target object within the preset measurement period by means of a first trained blood pressure calculation model based on the acquired electrocardiosignal and blood oxygen volume wave signal. ## Claims ### Claim 1 An electronic device, comprising: a processor configured to: acquire an electrocardiosignal and a blood oxygen volume wave signal of a target object within a preset measurement period; and determine beat-wise blood pressure values of the target object within the preset measurement period by means of a first trained blood pressure calculation model based on the acquired electrocardiosignal and blood oxygen volume wave signal. ### Claim 2 (depends on 1) The electronic device of claim 1, wherein the blood pressure values comprise a diastolic blood pressure and a systolic blood pressure, and wherein the processor is further configured to, after the beat-wise blood pressure values of the target object within the preset measurement period are determined: determine an error value corresponding to a diastolic blood pressure of each beat of the target object within the preset measurement period and an error value corresponding to a systolic blood pressure of each beat of the target object within the preset measurement period by means of a second trained blood pressure calculation model based on the acquired electrocardiosignal and blood oxygen volume wave signal; and correct the diastolic blood pressure of the corresponding beat determined by means of the first trained blood pressure calculation model based on the error value corresponding to the diastolic blood pressure of each beat of the target object within the preset measurement period so as to determine the corrected beat-wise diastolic blood pressures of the target object within the preset measurement period; and correct the systolic blood pressure of the corresponding beat determined by means of the first trained blood pressure calculation model based on the error value corresponding to the systolic blood pressure of each beat of the target object within the preset measurement period so as to determine the corrected beat-wise systolic blood pressures of the target object within the preset measurement period. ### Claim 3 (depends on 2) The electronic device of claim 2, wherein the processor is further configured to, before the beat-wise blood pressure values of the target object within the preset measurement period are determined based on the acquired electrocardiosignal and blood oxygen volume wave signal by means of the first trained blood pressure calculation model: perform a differential processing on the acquired blood oxygen volume wave signal to obtain a plurality of characteristic values of the blood oxygen volume wave signal; determine a pulse wave transmission time ‘PTT’ based on a maximum of a first derivative of the blood oxygen volume wave signal in the same cycle where the electrocardiosignal is located and an instant value of an R point of the electrocardiosignal; and perform a principal component analysis processing on the PTT and the plurality of characteristic values of the blood oxygen volume wave signal, and extract principal component factors that satisfy a preset condition to obtain a principal component factor matrix. ### Claim 4 (depends on 3) The electronic device of claim 3, wherein the processor is further configured to determine the beat-wise blood pressure values of the target object within the preset measurement period by means of the first trained blood pressure calculation model based on the acquired electrocardiosignal and blood oxygen volume wave signal by: inputting the principal component factor matrix to the first trained blood pressure calculation model to determine the diastolic blood pressure and the systolic blood pressure of the beat corresponding to the principal component factor matrix; and wherein the processor is further configured to determine the error value corresponding to the diastolic blood pressure of a beat of the target object within the preset measurement period and the error value corresponding to the systolic blood pressure of the beat of the target object within the preset measurement period by means of the second trained blood pressure calculation model based on the acquired electrocardiosignal and blood oxygen volume wave signal by: inputting the principal component factor matrix to the second trained blood pressure calculation model to determine the error value corresponding to the diastolic blood pressure and the error value corresponding to the systolic blood pressure of the beat corresponding to the principal component factor matrix. ### Claim 5 (depends on 1) The electronic device of claim 1, wherein the processor is further configured to, before the beat-wise blood pressure values of the target object within the preset measurement period are determined based on the acquired electrocardiosignal and blood oxygen volume wave signal by means of the first trained blood pressure calculation model: train an Elman neural network to obtain the first trained blood pressure calculation model. ### Claim 6 (depends on 5) The electronic device of claim 5, wherein the processor is further configured to train the Elman neural network to obtain the first trained blood pressure calculation model by: training the Elman neural network by taking a plurality of principal component factor matrixes acquired within a preset sampling period, and reference values of the diastolic blood pressures and reference values of the systolic blood pressures which are measured at the same beats as those for the respective principal component factor matrixes within the preset sampling period as training samples, so as to obtain the first trained blood pressure calculation model, wherein the principal component factor matrixes are obtained by performing differential processing and principal component analysis on the electrocardiosignals and blood oxygen volume wave signals. ### Claim 7 (depends on 2) The electronic device of claim 2, wherein the processor is further configured to, before the error value corresponding to the diastolic blood pressure of each beat of the target object within the preset measurement period and the error value corresponding to the systolic blood pressure of each beat of the target object within the preset measurement period are determined based on the acquired electrocardiosignal and blood oxygen volume wave signal by means of the second trained blood pressure calculation model: train a linear neural network to obtain the second trained blood pressure calculation model. ### Claim 8 (depends on 7) The electronic device of claim 7, wherein the processor is further configured to train the linear neural network to obtain the second trained blood pressure calculation model by: training the linear neural network by taking a plurality of principal component factor matrixes acquired within a preset sampling period and blood pressure error information corresponding to the respective principal component factor matrixes as training samples, so as to obtain the second trained blood pressure calculation model; wherein the blood pressure error information comprises: diastolic blood pressure error information and systolic blood pressure error information, wherein the diastolic blood pressure error information is a difference between the diastolic blood pressure output in the training process of the Elman neural network and the reference value of the diastolic blood pressure measured at the same beat; and the systolic blood pressure error information is a difference between the systolic blood pressure output in the training process of the Elman neural network and the reference value of the systolic blood pressure measured at the same beat. ### Claim 9 A blood pressure measuring method, comprising: acquiring an electrocardiosignal and a blood oxygen volume wave signal of a target object input within a preset measurement period; and determining beat-wise blood pressure values of the target object within the preset measurement period by means of a first trained blood pressure calculation model based on the acquired electrocardiosignal and blood oxygen volume wave signal. ### Claim 10 (depends on 9) The blood pressure measuring method of claim 9, wherein the blood pressure values comprise a diastolic blood pressure and systolic blood pressure, and the blood pressure measuring method further comprises: determining an error value corresponding to a diastolic blood pressure of each beat of the target object within the preset measurement period and an error value corresponding to a systolic blood pressure of each beat of the target object within the preset measurement period by means of a second trained blood pressure calculation model based on the acquired electrocardiosignal and blood oxygen volume wave signal; correcting the diastolic blood pressure of the corresponding beat determined by means of the first trained blood pressure calculation model based on the determined error value corresponding to the diastolic blood pressure of each beat of the target object within the preset measurement period so as to determine the corrected beat-wise diastolic blood pressures of the target object within the preset measurement period; and correcting the systolic blood pressure of the corresponding beat determined by means of the first trained blood pressure calculation model based on the determined error value corresponding to the systolic blood pressure of each beat of the target object within the preset measurement period so as to determine the corrected beat-wise systolic blood pressures of the target object within the preset measurement period. ### Claim 11 (depends on 10) The blood pressure measuring method of claim 10, wherein before determining beat-wise blood pressure values of the target object within the preset measurement period by means of a first trained blood pressure calculation model based on the acquired electrcardiosignal and blood oxygen volume wave signal, the blood pressure measuring method further comprises: performing a differential processing on the acquired blood oxygen volume wave signal to obtain a plurality of characteristic values of the blood oxygen volume wave signal; determining a pulse wave transmission time ‘PTT’ based on a maximum of a first derivative of the blood oxygen volume wave signal in the same cycle where the electrocardiosignal is located and an instant value of an R point of the electrocardiosignal; performing a principal component analysis processing on the PTT and the plurality of characteristic values of the blood oxygen volume wave signal; and extracting principal component factors that satisfy a preset condition to obtain a principal component factor matrix. ### Claim 12 (depends on 11) The blood pressure measuring method of claim 11, wherein determining beat-wise blood pressure values of the target object within the preset measurement period by means of a first trained blood pressure calculation model based on the acquired electrocardiosignal and blood oxygen volume wave signal comprises: inputting the extracted principal component factor matrix to the first trained blood pressure calculation model to determine the diastolic blood pressure and the systolic blood pressure of the beat corresponding to the principal component factor matrix; and wherein determining an error value corresponding to a diastolic blood pressure of each beat of the target object within the preset measurement period and an error value corresponding to a systolic blood pressure of each beat of the target object within the preset measurement period by means of a second trained blood pressure calculation model based on the acquired electrocardiosignal and blood oxygen volume wave signal comprises: inputting the extracted principal component factor matrix to the second trained blood pressure calculation model to determine the error value corresponding to the diastolic blood pressure and the error value corresponding to the systolic blood pressure of the beat corresponding to the principal component factor matrix. ### Claim 13 (depends on 9) The blood pressure measuring method of claim 9, further comprising: training an Elman neural network to obtain the first trained blood pressure calculation model. ### Claim 14 (depends on 13) The blood pressure measuring method of claim 13, wherein said training the Elman neural network to obtain the first trained blood pressure calculation model comprises: training the Elman neural network by taking a plurality of principal component factor matrixes acquired within a preset sampling period, and reference values of the diastolic blood pressures and reference values of the systolic blood pressures which are measured at the same beats as those for the respective principal component factor matrixes within the preset sampling period as training samples, so as to obtain the first trained blood pressure calculation model, wherein the principal component factor matrixes are obtained by performing differential processing and principal component analysis on the electrocardiosignals and blood oxygen volume wave signals. ### Claim 15 (depends on 10) The blood pressure measuring method of claim 10, further comprising: training a linear neural network to obtain the second trained blood pressure calculation model. ### Claim 16 (depends on 15) The blood pressure measuring method of claim 15, wherein said training the linear neural network to obtain the second trained blood pressure calculation model comprises: training the linear neural network by taking a plurality of principal component factor matrixes acquired within a preset sampling period and blood pressure error information corresponding to the respective principal component factor matrixes as training samples, so as to obtain the second trained blood pressure calculation model; wherein the blood pressure error information comprises: diastolic blood pressure error information and systolic blood pressure error information, wherein the diastolic blood pressure error information is a difference between the diastolic blood pressure output in the training process of the Elman neural network and the reference value of the diastolic blood pressure measured at the same beat; and the systolic blood pressure error information is a difference between the systolic blood pressure output in the training process of the Elman neural network and the reference value of the systolic blood pressure measured at the same beat. ### Claim at position 17 17-19. (canceled) ### Claim 20 A computer-readable storage medium on which a computer program is stored, wherein the computer program, when executed by a processor, causes the processor to: acquire an electrocardiosignal and a blood oxygen volume wave signal of a target object within a preset measurement period; and determine beat-wise blood pressure values of the target object within the preset measurement period by means of a first trained blood pressure calculation model based on the acquired electrocardiosignal and blood oxygen volume wave signal. ### Claim 21 (depends on 20) The computer-readable storage medium of claim 20, wherein the blood pressure values comprise a diastolic blood pressure and a systolic blood pressure, and wherein the computer program, when executed by the processor, further causes the processor to: determine an error value corresponding to a diastolic blood pressure of each beat of the target object within the preset measurement period and an error value corresponding to a systolic blood pressure of each beat of the target object within the preset measurement period by means of a second trained blood pressure calculation model based on the acquired electrocardiosignal and blood oxygen volume wave signal; correct the diastolic blood pressure of the corresponding beat determined by means of the first trained blood pressure calculation model based on the error value corresponding to the diastolic blood pressure of each beat of the target object within the preset measurement period so as to determine the corrected beat-wise diastolic blood pressures of the target object within the preset measurement period; and correct the systolic blood pressure of the corresponding beat determined by means of the first trained blood pressure calculation model based on the error value corresponding to the systolic blood pressure of each beat of the target object within the preset measurement period so as to determine the corrected beat-wise systolic blood pressures of the target object within the preset measurement period. ### Claim 22 (depends on 21) The computer-readable storage medium of claim 21, wherein the computer program, when executed by the processor, further causes the processor to: perform a differential processing on the acquired blood oxygen volume wave signal to obtain a plurality of characteristic values of the blood oxygen volume wave signal; determine a pulse wave transmission time ‘PTT’ based on a maximum of a first derivative of the blood oxygen volume wave signal in the same cycle where the electrocardiosignal is located and an instant value of an R point of the electrocardiosignal; and perform a principal component analysis processing on the PTT and the plurality of characteristic values of the blood oxygen volume wave signal, and extract principal component factors that satisfy a preset condition to obtain a principal component factor matrix. ### Claim 23 (depends on 22) The computer-readable storage medium of claim 22, wherein the computer program, when executed by the processor, further causes the processor to: input the principal component factor matrix to the first trained blood pressure calculation model to determine the diastolic blood pressure and the systolic blood pressure of the beat corresponding to the principal component factor matrix; and input the principal component factor matrix to the second trained blood pressure calculation model to determine the error value corresponding to the diastolic blood pressure and the error value corresponding to the systolic blood pressure of the beat corresponding to the principal component factor matrix.