Blood pressure algorithm
View Patent ↗A apparatus and method for non-invasively measuring blood pressure. Cuff pressure and oscillometric signals are generated. The signals are conditioned by a plurality of linear filters to remove artifacts and local trends. Rank-order filters are applied to the conditioned oscillometric signal to detect upper and lower envelops. The systolic blood pressure and diastolic blood pressure are calculated using a plurality of multidimensional threshold vectors.
1. A method for measuring non-invasive blood pressure from an oscillometric signal and a cuff pressure signal, said method implemented in a medical apparatus, comprising:
(a) calculating a pulse pressure signal by subtracting an upper and a lower envelope of said oscillometric signal based on rank-order filters applied to said oscillometric signals; and
(b) calculating a mean arterial pressure, a systolic blood pressure, and a diastolic blood pressure from said oscillometric signal, said cuff pressure signal, and a plurality of thresholds using a device with at least one processor.
2. The method of claim 1 , wherein said oscillometric signal is filtered with one or more frequency selective filters prior to the application of said rank-order filters, and said pulse pressure signal is further filtered using a median filter and a lowpass filter to eliminate artifacts.
3. The method of claim 2 , wherein calculating said mean arterial pressure is based on identifying an index value corresponding to a global maximum value in said pulse pressure signal and evaluating said cuff pressure signal at said index value.
4. The method of claim 3 , wherein calculating said systolic blood pressure and said diastolic blood pressure is based on evaluating said cuff pressure signal at index values corresponding to specific percent values of said maximum value in said pulse pressure signal, said percent values specified by said thresholds.
5. The method of claim 4 , wherein said plurality of thresholds are a function of the mean arterial pressure, resulting in a vector threshold with multiple scalar thresholds each corresponding to a different mean arterial pressure.
6. The method of claim 4 , wherein said plurality of thresholds are a function of an arm circumference, resulting in a vector threshold with multiple scalar thresholds each corresponding to a different arm circumference.
7. The method of claim 4 , wherein said plurality of thresholds are a function of a heart rate, resulting in a vector threshold with multiple scalar thresholds each corresponding to a different heart rate.
8. An apparatus for non-invasive measurement of blood pressuring, comprising:
(a) a signal acquisition circuit to acquire a cuff pressure signal using an inflatable cuff, and to generate an oscillometric signal from said cuff pressure signal;
(b) a memory to store said cuff pressure signal, said oscillometric signal, a plurality of processing steps; and a plurality of processing results; and
(c) a processor configured to perform the processing steps of:
(1) calculating a pulse pressure signal without the use of a beat detection algorithm by subtracting an upper and a lower envelope of said oscillometric signal based on rank-order filters applied to said oscillometric signals; and
(2) calculating a mean arterial pressure, a systolic blood pressure, and a diastolic blood pressure from said oscillometric signal, said cuff pressure signal, and a plurality of thresholds; and storing the results in said memory.
9. The apparatus of claim 8 , wherein (a) said upper and lower envelopes are calculated based on two rank-order filters applied to said oscillometric signal;
(b) said oscillometric signal is filtered with one or more frequency selective filters prior to the application of said rank-order filters, and said pulse pressure signal is further filtered using a median filter and a lowpass filter to eliminate artifacts;
(c) calculating said mean arterial pressure is based on identifying an index value corresponding to a global maximum value in said pulse pressure signal and evaluating said cuff pressure signal at said index value;
(d) calculating said systolic blood pressure and said diastolic blood pressure is based on evaluating said cuff pressure signal at two index values corresponding to two percent values of said maximum value in said pulse pressure signal, said percent values specified by said multidimensional threshold vectors; and
(e) said plurality of threshold vectors are a function of the mean arterial pressure, an arm circumference, and a heart rate.