System and method of a biosensor for detection of health parameters
A photoplethysmography (PPG) circuit obtains PPG signals at a plurality of wavelengths of light reflected from tissue of a user. A processing device generates parameters using the PPG signals to determine a glucose level in blood flow of the user. The parameters include one or more ratio values obtained using the plurality of PPG signals; a phase delay between the plurality of PPG signals; a correlation of phase shape between the plurality of PPG signals or a periodicity of one or more of the plurality of PPG signals.
1. A biosensor, comprising:
an optical circuit configured to obtain a plurality of PPG signals at a plurality of wavelengths reflected from tissue of a user, wherein the different wavelengths have varying penetration depths of tissue;
one or more processing circuits configured to:
determine a plurality of L values at a plurality of different wavelengths using the plurality of PPG signals, wherein a first L value of the plurality of L values is determined using a first PPG signal obtained at a first wavelength in a range of 380 nm-410 nm and a second L value of the plurality of L values is determined using a second PPG signal obtained at a second wavelength equal to or above 660 nm and a third L value of the plurality of L values is determined using a third PPG signal obtained at a third wavelength in a range of 510 nm-550 nm;
determine a plurality of R values using the plurality of L values, wherein a first R value is determined from a ratio of the first L value and the second L value and a second R value is determined from a ratio of the first L value and the third L value;
determine one or more other PPG parameters using the plurality of PPG signals; and
determine a glucose level in blood flow using the plurality of L values, the plurality of R values and the one or more other PPG parameters.
2. The biosensor of claim 1 , wherein the at least one processing circuit includes a first processing circuit configured to implement a regression neural network processing or a classifier neural network processing and wherein the plurality of L values, the plurality of R values and the one or more other PPG parameters are input into the first processing circuit to determine the glucose level.
3. The biosensor of claim 1 , wherein the one or more other PPG parameters include at least one of: a phase delay between a first PPG signal and a second PPG signal of the plurality of PPG signals, a correlation of phase shape between the first PPG signal and the second PPG signal of the plurality of PPG signals or a periodicity of first PPG signal or the second PPG signal.
4. The biosensor of claim 1 , wherein the processing circuit is further configured to determine a skin temperature and determine the glucose level using the plurality of L values, the plurality of R values, the one or more other PPG parameters and the skin temperature.
5. The biosensor of claim 1 , wherein the plurality of R values further includes:
a third R value determined from a ratio of the second L value and the third L value.
6. A biosensor, comprising:
an optical circuit configured to obtain a plurality of PPG signals at a plurality of wavelengths reflected from tissue of a user, wherein the different wavelengths have varying penetration depths of tissue;
one or more processing circuits configured to:
determine one or more R values using the plurality of PPG signals, wherein an R value is determined from a ratio of AC components of a first PPG signal obtained from a first wavelength and a second PPG signal obtained at second wavelength;
determine a plurality of PPG parameters using the plurality of PPG signals, wherein the plurality of PPG parameters includes: a phase delay between the first PPG signal and the second PPG signal of the plurality of PPG signals, a correlation of phase shape between the first PPG signal and the second PPG signal of the plurality of PPG signals and a periodicity of at least the first PPG signal or the second PPG signal; and
determine a glucose level in blood flow of the user using the one or more R values and the plurality of PPG parameters.
7. The biosensor of claim 6 , wherein the optical circuit is configured to obtain the first PPG signal at the first wavelength with a high absorption coefficient for nitric oxide (NO) in blood flow and the second PPG signal at the second wavelength with a low absorption coefficient for NO in blood flow.
8. The biosensor of claim 7 , wherein the one or more R values include:
a first R value obtained using the first PPG signal at the first wavelength with the high absorption coefficient for nitric oxide (NO) in blood flow and the second PPG signal at the second wavelength with a low absorption coefficient for NO in blood flow;
a second R value obtained using the first PPG signal at the wavelength with the high absorption coefficient for nitric oxide (NO) in blood flow and a third PPG signal at a third wavelength with a different penetration depth than the first and second wavelength; and
a third R value obtained using the third PPG signal at the third wavelength and the second PPG signal at the second wavelength with the low absorption coefficient for NO in blood flow.
9. The biosensor of claim 6 , wherein the processing circuit is further configured to determine the glucose level in blood flow of the user using the one or more R values, the plurality of PPG parameters and a plurality of L values, wherein the plurality of L values include:
a first L value determined using the first PPG signal at the first wavelength with a high absorption coefficient for nitric oxide (NO) in blood flow; and
a second L value determined using the second PPG signal at the second wavelength with a low absorption coefficient for NO in blood flow.
10. The biosensor of claim 6 , wherein the one or more processing circuits are further configured to determine a skin temperature and determine the glucose level using the one or more R values, the plurality of PPG parameters and the skin temperature.
11. The biosensor of claim 6 , wherein the one or more processing circuits implement a regression neural network processing or a classifier neural network processing to determine the glucose level.
12. The biosensor of claim 6 , wherein the one or more processing circuits are further configured to determine one or more digital health parameters using the plurality of PPG signals.
13. The biosensor of claim 12 , wherein the one or more digital health parameters includes a Vascular Health Index and wherein the processing circuit is configured to determine the Vascular Health Index using a measurement of a relative vasoconstriction of vessels during an insulin release event.
14. A biosensor, comprising:
an optical circuit configured to obtain a plurality of PPG signals at a plurality of wavelengths reflected from tissue of a user, wherein a first PPG signal is obtained at a first wavelength in a range of 380 nm-410 nm, a second PPG signal is obtained at a second wavelength equal to or above 660 nm, and a third PPG signal is obtained at a third wavelength in a range of 510 nm-550 nm;
a signal processing circuit configured to:
determine a plurality of L values at a plurality of different wavelengths using the plurality of PPG signals, wherein a first L value is determined by isolating an alternating current (AC) component of the first PPG signal and a second L value is determined by isolating an AC component of the second PPG signal and a third L value is determined by isolating an AC component of the third PPG signal;
determine a plurality of R values using the plurality of L values, wherein a first R value is determined from a ratio of the first L value and the second L value and a second R value is determined from a ratio of the first L value and the third L value; and
a neural network processing circuit that is configured to:
receive the plurality of L values and the plurality of R values as input; and
determine a glucose level in blood flow using the plurality of L values and the plurality of R values.
15. The biosensor of claim 14 , wherein the optical circuit is configured to obtain a third PPG signal at a third wavelength with a different penetration of depth from the first wavelength and the second wavelength.
16. The biosensor of claim 14 , wherein the neural network processing circuit is a regression neural network processing circuit or a classifier neural network processing circuit.
17. The biosensor of claim 16 , wherein the signal processing circuit is further configured to:
determine a plurality of PPG parameters using the plurality of PPG signals, wherein the plurality of PPG parameters include at least: a phase delay between the first PPG signal and the second PPG signal of the plurality of PPG signals, a correlation of phase shape between the first PPG signal and the second PPG signal of the plurality of PPG signals and a periodicity of at least the first PPG signal or the second PPG signal.
18. The biosensor of claim 17 , wherein the neural network processing circuit is further configured to determine the glucose level in blood flow using the plurality of L values, the plurality of R values, and the plurality of PPG parameters.
19. The biosensor of claim 14 , wherein the processing circuit is further configured to determine one or more digital health parameters using the plurality of PPG signals.
20. The biosensor of claim 19 , wherein the one or more digital health parameters includes a Vascular Health Index and wherein the processing circuit is configured to determine the Vascular Health Index using a measurement of a relative vasoconstriction of vessels during an insulin release event.