Description:Complete Specification:

The following specification describes and ascertains the nature of this invention and the and the manner in which it is to be performed.

Field of the invention
[0001] This invention is related to a control unit for assessing a quality of at least one signal generated in a bio-analyte device and a method thereof.

Background of the invention
[0002] The quality of the photoplethysmogram (PPG) signals plays an important role in any PPG-based device, as it drastically affects the outputs of any processes performed on it and any features derived from it. This is even more crucial in multi-wavelength devices as the complexity of the system is much higher, and a drop in quality of the PPGs in any wavelength can change the resultant outputs drastically. The baseline wander over time and the morphology/shape of the PPG pulses are the key measures of quality in the PPG signal. There have been algorithms created in the past to address the issue of baseline wander, by reducing it using methods such as baseline de-trending and adaptive noise cancelling, which are not only complex, but are also expensive in terms of time and hardware requirements. Similarly, there have been algorithms to assess the morphology of the PPG pulses but remain ineffective due to various reasons as known in the state of the art.

[0003] A EP patent application 3479763A1 discloses a system and method for PPG signal quality assessment is provided. The method includes obtaining a PPG signal captured using a testing device in real-time. The PPG signal is segmented into a first plurality of PPG signal samples such that length of each of the first plurality of PPG signal samples more than a threshold length. A signal sufficiency check (SSC) is performed for each first PPG signal sample of the first plurality of PPG signal samples to obtain at least a first set of PPG signal samples complying with the SSC. A set of features is extracted from the first set of PPG signal samples. Based on the set of features, each of the set of PPG signal samples is identified as one of a noisy signal sample and a clean signal sample using a plurality of Random Forest (RF) models created during the training phase. 
Brief description of the accompanying drawings
[0004] Figure 1 illustrates a control unit for assessing the quality of at least one signal in a bio-analyte device, in accordance with an embodiment of the invention; and
[0005] Figure 2 illustrates a flowchart for a method of assessing a quality of at least one signal in the bio-analyte device in accordance with the present invention.

Detailed description of the embodiments
[0006] Figure 1 illustrates a control unit for assessing the quality of at least one signal in a bio-analyte device, in accordance with an embodiment of the invention. The device 10 comprises a sensing element 14 for detecting the at least one signal 12, a signal processing module 16 and a quality check module 18. The control unit 10 removes artifacts from the detected at least one signal 12 by using the signal processing module 16 and transmits each pulse 13 of the at least one signal 12 through the quality check module 18. The control unit 10 calculates a ratio of magnitude of the difference from trough amplitudes and a peak height of each pulse 13 of the at least one signal 12 for determining a wander check factor.

[0007] The control unit 10 locates a key phase information on each pulse 13 of the at least one signal 12 by identifying multiple peaks and zero-crossings in a first derivative and a second derivative of the pulse 13 and constructs a standard piece-wise linear first derivative by maintaining the key phase information. The control unit 10 generates a template pulse 13(a) by integrating the piece wise linear first derivate by producing a smooth quadratic graph and assesses the quality of each pulse 13 of the at least one signal 12 by comparing a correlation coefficient determined between the created template pulse 13(a) and the each pulse 13 of the at least one signal 12.

[0008] Further the construction and the working of the components of the device 11 and the control unit 10 are explained in detail. The device 10 being the bio-analyte device is used to determined at least one human parameters like a Hemoglobin, blood pressure, any one of the body parameters like thyroid levels, blood sugar level and the like. It is to be understood that the parameter that is determined by using the bio-analyte device 10 is not limited to the above mentioned one’s but can be of any other parameters that is known for a person skilled in the art. The bio-analyte device 10 comprises at least one signal light source (not shown) having a corresponding wavelength and a sensing element 14 for detecting/sensing the at least one signal 12 generated from the light source having a wavelength. For instance, the bio-analyte device 11 according to one embodiment of the invention comprises four light sources having corresponding four wavelengths.

[0009] The sensing element 14 provided in the device detects/senses each signal 12 from each of the light source. I.e.., the sensing element 14 detects four signals during an operational state of the device 11. And each signal 12 comprises multiple pulses 13. For better understanding of the invention, if each signal 12 comprises ten pulses 13, the control unit 10 have to assess the quality of forty (ten* four) pulses 13 during the operational state of the device 10.

[0010] According to one embodiment of the invention, the at least one signal 12a is a photoplethysmogram signal detected using any one of the following sensing element 14 comprising a multi-wavelength sensing element and a single wavelength sensing element as mentioned above. The type of sensing element 14 is not explained in detail as it is known in the state of the art. The quality check module 18 comprises two units 18(a) & 18(b), a first quality unit 18(a) for determining a baseline wander factor and a second quality unit 18(b) for determining a pulse morphological factor. The first quality unit 18(a) determines the baseline wander/wander check factor by comparing the calculated ratio with a threshold value.

[0011] The baseline wander check includes calculating the ratio of the magnitude of the difference in trough amplitudes and the peak height as disclosed. This ratio gives the tilt of the base of the each pulse 13 normalized with the peak height, which acts as an effective indicator of baseline wander in each pulse 13. This ratio is compared with the threshold value and based on that, the pulse 13 is either accepted or rejected.., if the calculated ratio is lower than the threshold value then, the pulse 13 is accepted and if the ratio is higher than the threshold value, the pulse 13 is rejected. The accepted pulses are passed through the pulse morphology check module/second quality unit 13(b) to assess the morphology of the pulse.

[0012] The second quality unit 18(b) of the quality check module locates the key points by maintaining the key phase information in each pulse 13. The second quality unit 18(b) locates said key phase information as a systolic peak, a diastolic peak, and a dicrotic notch in each of the pulse 13, by locating multiple peaks and zero-crossings in the first derivative and in the second derivative of the pulse 13. The control unit 10 sets amplitudes of the constructed standard piece -wise linear first derivative such that, when integrated, amplitudes of the each pulse 13 and the template pulse 13(a) match at the key points. The constructed template pulse 13(a) by integrating the standard piece-wise linear first derivative produces a smooth quadratic curve.

[0013] According to one embodiment of the invention, the correlation coefficient is a Pearson’s correlation coefficient. The correlation coefficient is compared a predefined threshold value, for assessing the quality of the each pulse 13 of the at least one signal 12, is either accepted or rejected if the coefficient is higher or lower than the threshold respectively. The signal 12 is accepted if the coefficient is higher than the predefined threshold value and the signal 12 is rejected by the control unit 10 if the coefficient is lower than the predefined threshold value. In the figure 1, the color curve represents the template pulse 13(a) and the blue curve represents the original pulse 13 of the at least one signal 12.

[0014] Figure 2 illustrates a flow chart for a method of assessing a quality of at least one signal in the bio-analyte device in accordance with the present invention. The device 11 comprises a sensing element 14, a control unit 10 having a signal processing module 16 and a quality check module 18. The method comprises the following steps. In step S1, artifacts from the detected at least one signal 12 are removed by using said signal processing module 16. In step S2, each pulse of the at least one signal 12 is transmitted through the quality check module 18.

[0015] In step S3, a ratio of magnitude of the difference from trough amplitudes and a peak height of the each pulse 13 of the at least one signal 12 is calculated for determining a wander check factor. In step S4, a key phase information on the each pulse 13 of the at least one signal 12 is located, by identifying multiple peaks and zero-crossings in a first derivative and a second derivative of the each pulse 13. In step S5, a standard piece-wise linear first derivative is constructed by using and maintaining the key phase information. In step S6, a template pulse 13(a) is generated by integrating the piece wise linear first derivate by producing a smooth quadratic graph. In step S7, the quality of the each pulse 13 of the at least one signal 12 is assessed by the control unit 10 by comparing a correlation coefficient determined between the created template and the each pulse 13 of the at least one signal 12.

[0016] The above method is explained in detail by referring the signal as the PPG signal 12. The PPG signal 12 is a time series photoplethysmogram (PPG) signal obtained from the sensing element 14 /light source. Based on the number of wavelengths of light used, the output of the sensing element 14 will be plurality of PPG signals 12 each corresponding to a particular wavelength. Each PPG signal 12 is a set of PPG pulses 13 consisting of one systole and one diastole phase of a heartbeat. The device 11 consists of a multi-wavelength/single-wavelength PPG sensing element 14 that senses/detects the PPG signals 12, the signal processing module 16, and the PPG signal quality check module 18 that assesses the quality and outputs part of the signal 12 (PPG pulses 13), that are of acceptable quality. The good quality signal 12 is further processed and analyzed to give a health-related metric. The control unit 10 asses the PPG signal 12 quality in terms of baseline wander factor and in terms of pulse morphological factor.

[0017] The control unit 10 removes artifacts using any one of the signal processing techniques and reduces the extent of baseline wander followed by peak-trough detection, trough and pulse alignment, and finally isolation of each pulse 13. Each isolated pulse 13 in the PPG signal 12 of a particular wavelength is passed through a signal quality check module 18 which consists of two quality units (18(a) & 18(b)). The first quality unit 18(a) being related to baseline wander check and the second quality unit 18(b) being related to pulse morphology check.

[0018] The baseline wander is a low frequency noise, and the PPG signal 12 being a low frequency signal, baseline wander cannot be eliminated without losing significant information in the PPG signal 12. Therefore, the control unit 10 filters the PPG signal 12 and the baseline wander factor is reduced without affecting the PPG signal 12 features significantly. The extent of the baseline wander factor that exists post-filtering is assessed using the ratio. The first quality unit 18(a) calculates the ratio of the magnitude of the difference in trough amplitudes and the peak height for the determining the baseline wander check factor. This ratio gives the tilt of the base of the PPG pulse 13 normalized with the peak height, which acts as an effective indicator of baseline wander in each pulse 13. This ratio is compared with the threshold value and based on the comparison, the pulse 13 is either accepted or rejected. If the ratio is lower than the threshold, the pulse 13 is accepted and if the ratio is higher than the threshold value, the pulse 13 is rejected.

[0019] The accepted pulses 13 are passed through the second quality unit 18(b) which is related for assessing the morphology of the pulse. One way to assess the morphological quality of the PPG signal 12 is to assess the pulses 13 that compose it. The assessment of the morphological quality of the PPG pulse 13, is by template matching, which is when the pulse 13 is correlated with a predetermined template pulse to check how morphologically close the two of them are. But having a single template to match is ineffective in assessing the quality of PPGs 12. This is because, in different subjects, the component waves of a PPG pulse (mainly the systole and the diastole) are located at different positions within the pulse i.e., the phase of the waves in the pulse is different, as the phase is dependent on different factors such as arterial resistance, heart rate, etc.

[0020] In order to improve the accuracy on the template matching, the control unit 10 generates a template pulse 13(a) for each pulse 13 with the same phase information as the original pulse 13. The quality assessment using the morphological check is explained in detail. The control unit 10 locates key phase information or the key points such as the systolic peak, the diastolic peak, and the dicrotic notch in the original pulse 13. It is to be understood that the pulse 13 of the at least one signal 12 is referred as the original pulse 13 in some parts of this document. The control unit 10 locates the key phase information by identifying the peaks and zero-crossings in its first derivative and second derivative of the pulse 13.

[0021] After locating the key phase information, the control unit 10 constructs the standard piece-wise linear first derivative by maintaining the key phase information that was determined above. The amplitudes of the constructed first-derivative of the each pulse 13 are set using mathematical equations, such that when integrated, the amplitudes of the original pulse 13 and template pulse 13(a) are matched at the key points. The control unit 10 constructs the template pulse 13(a) by integrating the standard piece-wise linear first derivative, producing a smooth quadratic curve. The control unit 10 calculates the correlation coefficient (which according to one embodiment of the invention is a Pearson’s correlation coefficient) between the template pulse 13(a) and the original pulse 13.

[0022] The correlation coefficient is compared with a predetermined threshold value, and the pulse 13 is either accepted or rejected based on the comparison. The control unit 10 accepts the pulse if the calculated coefficient is higher than the predetermined threshold value and rejects the pulse 13 if the coefficient is less than the threshold value. Based on the number of accepted pulses 13 synchronized across all the wavelengths (in case of a multi-wavelength PPG system), the sample of PPG signals 12 are classified as perfect, accepted, or rejected by the control unit 10.

[0023] With the above disclosed method of assessing the quality of the at least one signal 12, different key points in the PPG pulse 13 can be chosen. The method can be implemented as a stand-alone entity or as an integrated method in any of the systems/devices 11 that have the signal which needs the quality assessment. The above disclosed method can be used in the real-time or in the post -signal acquisition time. The sensing element 14 can be used for single wavelength or multiple wavelengths depending on the requirement.

[0032] Embodiments explained in the description above are only illustrative and do not limit the scope of this invention. Many such embodiments and other modifications and changes in the embodiment explained in the description are envisaged. The scope of the invention is only limited by the scope of the claims
, Claims:We Claim:

1. A control unit (10) for assessing a quality of at least one signal (12) generated in a bio-analyte device (11), said device (11) comprises a sensing element (14) for detecting said at least one signal (12); a signal processing module (16) and a quality check module (18);

said control unit (10) adapted to:
- remove artifacts from said detected at least one signal (12) by using said signal processing module (16);
- transmit each pulse (13) of said at least one signal (12) through said quality check module (18);
characterized in that:
- calculate a ratio of magnitude of the difference from trough amplitudes and a peak height of said each pulse (13) of said at least one signal (12) for determining a wander check factor;
- locate a key phase information on said each pulse (13) of said at least one signal (12) by identifying multiple peaks and zero-crossings in a first derivative and a second derivative;
- construct a standard piece-wise linear first derivative by using and maintaining said key phase information;
- generate a template pulse (13(a)) by integrating said piece wise linear first derivate by producing a smooth quadratic graph;
- assess said quality of said each pulse (13) of said at least one signal (12) by comparing a correlation coefficient determined between said created template and said each pulse (13) of said at least one signal (12).

2. The control unit (10) as claimed in claim 1, wherein said at least one signal (12) is a photoplethysmogram signal detected using any one of the following sensing element (14) comprising a multi-wavelength sensing element and a single wavelength sensing element.

3. The control unit (10) as claimed in claim 1, wherein said quality check module (18) comprises two units (18(a) & 18(b)), a first quality unit (18(a)) for determining a baseline wander factor and a second quality unit (18(b)) for determining a pulse morphological factor.

4. The control unit (10) as claimed in claim 3, wherein said first quality unit(18(a)) determining said baseline wander/wander check factor by comparing said calculated ratio with a threshold value.

5. The control unit (10) as claimed in claim 1, wherein said second quality unit (18(b)) adapted to locate said key points by maintaining said key phase information in said each pulse (13).

6. The control unit (10) as claimed in claim 5, wherein said second quality unit (18(b)) locates said key phase information as a systolic peak, a diastolic peak, and a dicrotic notch in each of said pulse (13), by locating multiple peaks and zero-crossings in said first derivative and in said second derivative of said each pulse (13).

7. The control unit (10) as claimed in claim 1, wherein said control unit (10) sets amplitudes of said constructed standard piece -wise linear first derivative such that, when integrated, amplitudes of said each pulse (13) and said template pulse (13(a)) match at said key points.

8. The control unit (10) as claimed in claim 1, wherein said correlation coefficient is compared a predefined threshold value, for assessing said quality of said each pulse (13) of said at least one signal (12), is either accepted or rejected if the coefficient is higher or lower than the threshold respectively.

9. The control unit (10) as claimed in claim 8, wherein said at least one signal (12) is accepted if said coefficient is higher than said predefined threshold value and said at least one signal (12) is rejected by said control unit (10) if said coefficient is lower than said predefined threshold value.

10. A method of assessing the quality of at least one signal (12) in a bio-analyte device (11), said device (11) comprising a sensing element (14), a control unit (10) having a signal processing module (16) and a quality check module (18), said method comprising the steps of:
- removing artifacts from said detected at least one signal (12) by using said signal processing module (16);
- transmitting each pulse (13) of said at least one signal (12) through said quality check module (18);

characterized in that:

- calculating a ratio of magnitude of the difference from trough amplitudes and a peak height of said each pulse (13) of said at least one signal (12) for determining a wander check factor;
- locating a key phase information on said each pulse (13) of said at least one signal (12) by identifying multiple peaks and zero-crossings in a first derivative and a second derivative;
- constructing a standard piece-wise linear first derivative by using and maintaining said key phase information;
- creating a template pulse (13(a)) by integrating said piece wise linear first derivate by producing a smooth quadratic graph;
- assessing said quality of said each pulse (13(a)) of said at least one signal (12) by said control unit (10) by comparing a correlation coefficient determined between said created template pulse (13(a)) and said each pulse (13) of said at least one signal (12).