4. PREAMBLE TO THE DESCRIPTION
COMPLETE SPECIFICATION:
The following specification particularly describes the invention and the Manner in which it is to be performed:
5.BACKGROUND OF THE INVENTION
A.Field of invention
This invention relates to a method and apparatus for measuring Blood pressure in humans by acquiring Oscillometric waveforms from Human body using the help of Blood pressure cuff, signal conditioning and processing the acquired waveform using plurality of Signal conditioning techniques, applying curve fit techniques and taking first order derivative to get the Systolic and Diastolic Blood Pressure values.
B.Description of related art
In present day and age due to both the lack of exercise and dietary habits Health ailments like high blood pressure, heart problems, stress and other related health conditions are on the rise.
For Blood pressure measurement two methods are employed one of which is an invasive measurement using a catheter inserted into the artery of the patient in a clinical environment by trained physicians and other method involves the use of tying an occluding bladder called as a cuff over the brachial artery and measuring the blood pressure. The principle of using an occluding bladder is called as Oscillometric Blood pressure measurement where a pressure sensor is used to sense the minute pressure changes due to change in arterial pressure which is transferred to the cuff. The oscillations start with minimal to a point of maximum . called as Mean Arterial Pressure (MAP) and drops back to minimal as the pressure is constantly reduced from an occluded bladder.
There has been many attempts to provide algorithms for Blood pressure monitoring and as referred in U.S. Pat. No. 4,869,261 which uses fixed ratios from MAP to calculate blood pressure, US. Pat No. 5,704,362 performing curve fitting and then

using fixed ratios to calculate blood pressure, US. Pat No. 5,579,776 using Step Deflation Solenoid and fixed ratio's to measure blood pressure, U.S. Pat. No. 6,413,223 Bl which models arterial blood flow and uses decomposition to calculate blood pressure, U.S. Pat. No. 6,893,403 uses a predefined envelopes and compares the resultant values with the reference values and finally uses a fixed threshold value, U.S. Pat. No. 7,927,283 using Rank order filters and fixed threshold vectors. The common denominator between all these patents is the usage of fixed ratio from a reference point to calculate the Blood pressure values.
This invention removes the need for a fixed ratio to find the Systolic and Diastolic points and proposes a novel approach where in a polynomial equation is generated based on the processed oscillometric waveform at run time and the second derivative of the polynomial equation is used to find the Systolic and Diastolic points thus eliminating the need for fixed ratios.
6.SUMMARY OF THE INVENTION
The current invention is intended to be an integrated, portable, wireless health tracker which interface's to a mobile or desktop computing device using a wireless interface to send and receive the collected and compute data. In its exemplary embodiment the invention includes a Blood pressure Cuff, Motor and Solenoid assembly for inflating and deflating the cuff and a microcontroller with wireless communication to transmit the received data.
In an exemplary embodiment of the device the number of oscillatory waveform received by the systems is fixed at a specific number irrespective of heart rates. This is achieved by the software running on the invention continuously scanning for the heart rate of the user and controlling the resultant rate of deflation using a linear deflation solenoid.
Another exemplary embodiment is to use the resultant signal and take the Fast Fourier Transform (FFT) the same and find the frequency of the waveform. This frequency is then converted to time domain to find the heart rate.
Another exemplary embodiment is to achieve a linear rate of deflation based on Heart rate of the individual by using a Proportional-Integral-Derivative (PID) controller by generating an expected waveform using a Digital to Analog circuit and convertor to maintain linear rates of deflation based on the detected heart rate..
Yet another exemplary embodiment the present invention provides a method of arriving at the signal envelope by concatenating the received signal, with its mirror image to form a larger signal double than its original length. The resultant signal is fed through an envelope detector like Hilbert transform and output of the envelope detector is then filtered using a Band-pass filter. The resultant signal is truncated to its original non-mirrored signal length.

Another exemplary embodiment is to take the peaks and troughs of the processed signal and fit a Guass curve over the resultant peaks using the error value which is a measure the closeness of the fitted curve against the input peaks. The peaks are then removed at random to achieve a resultant error value of less than 10% when compared to the original is used.
Yet another exemplary embodiment is use the Curve fitted graph as input and a polynomial equation is derived based the same. The resultant coefficients of the polynomial equation is fed into polynomial differentiation and the results plotted.
In another exemplary embodiment of the invention is to use the least value obtained by the polynomial differential and map the equivalent pressure value to the Systolic point and use the maximum value obtained by the derivative to the Diastolic point.

7. DESCRIPTION
A new approach to measure Non Invasive Blood Pressure is described in the below sections with the design and methods used in the proposed invention.
From Figure 1 shows an exemplary embodiment of the invention where the invention has an Analog to Digital cohvertor and Digital to Analog convertor connected to a computing device like a microprocessor or microcontroller. The Digital to Analog convertor is interfaced with PID controller which gives current proportional to the deflation rate to the linear solenoid. The solenoid is in-turn interfaced with the Blood Pressure cuff and another opening of the Blood Pressure cuff is interfaced with the Piezo Electric pressure sensor. The pressure sensor gives output voltage corresponding to the pressure in the cuff which in-turn is fed into the Analog to Digital convertor for processing. Using this approach the software in the microprocessor/microcontroller keeps monitoring the pressure release rate in mm/Hg. Once a few peaks are received the invention calculates the heart rate and decides on the pressure rate based on a pre-defined lookup available with the software of invention. This data is fed into the Digital to Analog circuitry which in-turn generates an analog voltage signal with a fixed slope. This signal is fed into the input of the PID controller which takes the current pressure and the required pressure hence maintaining fixed rates of deflation.
Figure 2 shows an exemplary embodiment of the device in operation to measure the Blood Pressure from the ADC inputs values. The signal is first passed through a series of linear band pass filters to filter the high frequency components above 20 Hz and low frequency components below 1 Hz. Figure 3 shows the result of the bandpass . filtering with the filtered signal showing the Oscillometric components plotted with the signal received from the ADC showing the actual pressure. This signal is fed into an envelope detector to extract the signal envelope of the waveform.
Figure 4 shows another exemplary embodiment of the method in using Guass-Newton method of curve fitting to fit a model to the data obtained by the envelope detector by using sum of squares method to reduce the error rate between the Gauss-newton model result and the actual points. The residual error between the Guass-newton method and the actual points is defined in-terms of
ri = Vi - fwifi) where r is the error rate, x is the input value, y is the output value
and P = (A-AJ is the parameters of the model. The method uses the gauss newton method to minimize the error rate Y to attain convergence. In the event of convergence not happening, the method takes the error rate which is the most minimal that is the point where the convergence was maximum and uses the obtained parameters. The program is stopped when convergence value of more than 90% or an error rate of less than 10% is achieved by the Gauss-Newton curve fit method.
Figure 5 shows the corrected peak heights of the oscillometric waveform based on the results obtained by the Gauss-Newton method. The new resultant peak heights

are fed to create a Vandermonde matrix also called as Alternant matrix as shown below,
Where X is the peak heights which are generalized and the residual taken, P is the polynomial coefficients and y is the output vector. This resultant matrix is solved to get a polynomial equation for the peak height values.
Figure 6 shows yet another exemplary embodiment where a polynomial differentiation is taken on the resultant polynomial equation obtained by curve as per the below equation and the results plotted.
k fxi =7p x, .
Figure 7 shows the second order polynomial derivative of the equation plotted against the pressure axis. Figure 8 shows another exemplary embodiment where the minimal value to left side of the peak point achieved by the derivative shows the Diastolic point which in the example given is at the pressure value of 78 mm/Hg and the Systolic point is showcased in Figure 9 showing the Systolic value at 120 mm/Hg which is minimal point to the right side of the peak point. The peak point is the representation of the Mean Artery Pressure (MAP). The values Systolic, Diastolic and heart rate are given to the display system or transmitted to the mobile or desktop computing unit and stored.
It is understood by those skilled in the art that details regarding some specific exemplary embodiments described above was provided only for clarification in the overall understanding and the scope of the present invention and the invention is not limited only to examples and methods described and hence various changes on size, form or methodologies could be made without deviating from the real scope of the invention and the claims mentioned below.

TITLE: NOVEL NONINVASIVE BLOOD PRESSURE MONITORING SYSTEM 8.BRIEF DESCRIPTION OF THE DRAWINGS
Figure 1 shows the block diagram of the apparatus.
Figure 2 shows the flow diagram of the method in its different stages from the input of the raw value to the final result display.
Figure 3 show the raw readings received from the ADC from maximum pressure to minimum in mm/Hg and the resultant value of filtering.
Figure 4 shows the output of the envelope detector.
Figure 5 shows the result of applying Gaussian curve fit on the envelope detector output.
Figure 6 shows the result of taking polynomial derivative on the curve which shows the maxima and the minima of the graph.
Figure 7 shows the output of the Second order derivative of Figure 7 plotted against the pressure values.
Figure 8 shows the zoomed in image of the Diastolic section of the graph.
Figure 9 shows the zoomed image of the Systolic section of the graph.

9.CLAIMS We Claim that:
1. A method of determining Blood pressure namely Systole, Diastole values using curve fit and polynomial derivatives.
2. A method of maintaining linear air flow rate from blood pressure cuff using PID controller a linear deflation solenoid.
3. A method of detecting and setting fixed deflation rates for various heart rates using Fast Fourier Transform to find the heart rates.
4. A method of claim 2 wherein the number of oscillations or oscillometric pulses received by the blood pressure apparatus is fixed within a certain threshold bound irrespective of heart rates.
5. A method claim 1 wherein the received signal is curve fitted using Gauss-Newton method and removing the outlier points.
6. A method of claim 5 where in a polynomial equation is derived based on the curve fit values using Vandermonde matrix method to find the coefficients.
7. A method of claim 6 wherein polynomial derivative is taken to find the Mean Arterial pressure (MAP) as the peak point of the signal obtained by the polynomial derivative.
8. A method of claim 7 wherein the Systole and Diastole points are computed by taking the minimal values obtained by the polynomial derivative of the signal.